≡ Menu

KOL467 | Discussing AI and IP with Juani from Argentina

Play

Kinsella on Liberty Podcast: Episode 467.

I was asked one Juani (@witheredsummer), an Argentinian Voluntarist, on Twitter and then later via email, to review his draft article “Ideas Are Free: A Case Against Intellectual Property” (text below) (I note that perhaps ironically, this title is almost identical to one of my own previous publications/speeches, Ideas are Free: The Case Against Intellectual Property: or, How Libertarians Went Wrong). I told him it was too long for me to fisk, got Grok to analyze it (see below), and told him to read up on some of my work and review the analysis, so that we could have a more productive conversation about it, and which I could also at least record for my podcast. This is our discussion.

In the end, he basically identifies many problems with and absurdities with IP law, which he’s right about and which I’ve mentioned, and comes up with some proposals for IP reform that would reduce its harm—again, which he’s right about and which I and others have also proposed, but also which are unlikely to be adopted by those infested with the IP mind-virus.

GROK SHOWNOTES:

[0:00-15:00] In this episode of the Stephan Kinsella podcast (KOL467), host Stephan Kinsella engages with Juani from Argentina to discuss intellectual property (IP) and its implications, particularly in the context of artificial intelligence (AI). Juani, a programmer and self-described libertarian, shares his consequentialist critique of IP, influenced by Kinsella’s work and his own essay, “Ideas for Free: A Case Against Intellectual Property.” The conversation begins with Juani outlining his concerns about IP’s practical harms, such as its impact on innovation and culture, and transitions into a discussion about AI. They explore how AI’s reliance on vast datasets, often containing copyrighted material, raises legal questions about copying, authorship, and derivative works. Kinsella highlights the tension between copyright law and AI development, noting that current laws could stifle AI’s potential by limiting data access or imposing costly licensing requirements.

[15:01-1:37:33] The discussion deepens into specific examples of IP’s negative effects, including cultural erasure through Disney’s sanitized retellings of historical figures like Pocahontas and Mulan, and the economic burdens of pharmaceutical patents, which restrict access to life-saving drugs in poorer nations. Juani argues that IP acts as a form of censorship and reinforces wealth disparities, citing cases like patent trolling and the high cost of educational materials in developing countries. Kinsella agrees, emphasizing that IP distorts markets and innovation, and suggests that copyright’s harm to AI development may rival its threat to internet freedom. Toward the end, Juani proposes a reform to replace IP exclusivity with a decaying royalty system, which Kinsella views skeptically, arguing that entrenched IP proponents will resist any reduction in rights. The episode concludes with reflections on piracy, the success of platforms like Steam, and the cultural shift toward viewing copying as less harmful, signaling a potential change in public perception of IP.

 

Youtube Transcript and detailed Grok shownotes below.

Related:

Juani sent me these links:

  • On Milei’s reforms in terms of copyright; in Spanish; in Spanish
  • As for that one story of IP in Sybaris, I mentioned it comes from Deipnosophistae, which is an Ancient Greek text. I do owe you the exact reference though, but it is supposed to be about Chapter 20 of Book III, or section 98e of Book 3, or so is cited elsewhere.
  • And the book I recommended is a collection of excerpts from the Zhuangzi, many of which echo a Libertarian, pro-spontaneous order message all the way back in the 3rd century BCE. I own a different edition of the book, but the contents should be the same: Thomas Merton, The Way of Chuang Tzu

Detailed Grok Shownotes

Bullet-Point Summary for Show Notes with Time Block Descriptions
Summary Overview: In episode KOL467, Stephan Kinsella and Juani from Argentina explore the multifaceted issues surrounding intellectual property (IP), with a focus on its intersection with artificial intelligence (AI), cultural impacts, and economic consequences. Their discussion critiques IP’s philosophical and practical shortcomings, using examples from AI lawsuits, Disney’s cultural narratives, and pharmaceutical patents. Juani advocates for reform, while Kinsella emphasizes the systemic flaws of IP, predicting resistance to change but noting technology’s role in shifting attitudes toward piracy.
0:00-15:00: Introduction and AI-IP Tensions
  • Description: Kinsella introduces Juani, a programmer from Argentina, and sets the stage for discussing Juani’s essay critiquing IP. They delve into how AI’s data-heavy training processes conflict with copyright law, exploring issues of authorship, derivative works, and fair use.
  • Summary:
    • [0:00-0:48] Kinsella opens the podcast, noting it as KOL467, and introduces Juani, who has written a 25-page draft on IP, influenced by Kinsella’s work.
    • [0:49-2:54] Juani describes his libertarian, consequentialist perspective on IP, shaped by personal thinking and Kinsella’s writings. He identifies as a voluntarist with Austrian economic sympathies.
    • [3:01-5:19] They discuss Argentine reforms under Milei, including the removal of music royalty regulations, which caused controversy among mainstream musicians.
    • [5:20-9:26] Juani introduces his essay’s focus on AI and IP, highlighting lawsuits like Authors Guild vs. Open AI, questioning how copyright applies to AI-generated content.
    • [9:27-15:00] Kinsella clarifies authorship issues (e.g., the monkey selfie case) and explains copyright’s scope, including literal copying and derivative works. He notes that AI training involves copying that may violate copyright, potentially hobbling AI development.
15:01-30:00: AI’s Practical Challenges and Broader IP Harms
  • Description: The conversation shifts to the practical implications of copyright on AI, including the infeasibility of filtering massive datasets and the chilling effect of lawsuits. They also touch on non-IP concerns, like AI ethics and state regulation risks.
  • Summary:
    • [15:01-19:58] Juani details how AI training datasets inevitably include copyrighted material, making compliance with copyright law impractical. Kinsella agrees, noting that regulators prioritize enforcement over feasibility.
    • [20:00-24:58] They discuss legal ambiguities around AI-generated works, with Kinsella suggesting that courts may adapt by presuming access to copyrighted material, shifting burdens of proof.
    • [25:00-30:00] Juani expresses concern about state regulation of AI on ethical grounds, fearing it could stifle innovation. Kinsella counters that copyright already regulates AI, amplifying its harm by limiting data access.
30:01-45:00: Cultural and Economic Impacts of IP
  • Description: The focus turns to IP’s cultural and economic consequences, with examples like Disney’s control over Mickey Mouse and sanitized historical narratives. They discuss how IP enables censorship and reinforces wealth disparities.
  • Summary:
    • [30:01-34:30] Kinsella argues that copyright’s distortion of AI may be its greatest harm, surpassing its threat to internet freedom. Juani predicts lawsuits from copyright holders will hinder AI progress.
    • [34:31-39:00] They critique IP’s incompatibility with AI, as it struggles to define authorship when machines generate content. Juani fears state surveillance using AI, enabled by lax self-regulation.
    • [39:01-45:00] Juani highlights Disney’s extended copyright on Mickey Mouse and attempts to patent a musical scale, illustrating IP’s absurdity. Kinsella cites the “Happy Birthday” lawsuit as an example of vague, bullying IP enforcement.
45:01-1:00:00: Cultural Erasure and Creative Freedom
  • Description: Juani elaborates on IP as cultural erasure, using Disney’s Pocahontas and Mulan as case studies. They discuss how IP restricts creative freedom, particularly for musicians and marginalized voices.
  • Summary:
    • [45:01-50:00] Juani explains how Disney’s versions of Pocahontas and Mulan distort historical and cultural narratives, erasing indigenous and Chinese identities. Kinsella notes the legal risks of challenging Disney’s narratives.
    • [50:01-55:00] Juani, a musician, argues that music thrives without copyright in his local DIY scene. Kinsella reinforces that historical inventions and scientific discoveries occurred without IP protection.
    • [55:01-1:00:00] They debunk the myth that IP drove the Industrial Revolution, citing the Venetian patent system’s failure to make Venice a technological hub. Kinsella references ancient examples of monopoly privileges, tying IP to flawed labor-based property theories.
1:00:01-1:15:00: Censorship and Patent Trolling
  • Description: The discussion explores IP as censorship, particularly when publishers control artists’ work. They address patent trolling and pharmaceutical patents’ deadly consequences, emphasizing global inequities.
  • Summary:
    • [1:00:01-1:05:35] Juani describes how publishers owning copyrights can censor artists with controversial views, citing Ted Nugent and Morrissey. Kinsella shares a fictional example from “The Studio” and a real case of a DMCA strike against DragonForce’s Herman Li.
    • [1:05:36-1:10:00] Juani introduces his “PPP doctrine” (patent, protect, perpetuate), arguing that IP creates a prisoner’s dilemma. Kinsella clarifies that only authors can copyright, but parallel invention in patents poses risks.
    • [1:10:01-1:15:00] Juani details patent trolling (e.g., NTP vs. Research in Motion) and biopiracy in India, which exacerbate global wealth gaps. Kinsella confirms that Western industries benefit disproportionately from IP.
1:15:01-1:37:33: Education, Reform, and Piracy
  • Description: The final segment addresses IP’s impact on education, tragic cases like Aaron Swartz, and Juani’s reform proposal. They conclude with optimism about technology and piracy reshaping IP perceptions.
  • Summary:
    • [1:15:01-1:20:18] Juani shares how copyright limits educational access in Argentina, forcing reliance on pirated materials. Kinsella recounts Aaron Swartz’s suicide due to copyright charges for sharing research papers.
    • [1:20:19-1:25:00] Juani proposes a decaying royalty system to replace IP exclusivity, allowing use with payments that decrease over time. Kinsella doubts its feasibility due to IP proponents’ resistance.
    • [1:25:01-1:30:00] They discuss public attitudes toward piracy, noting that technology (e.g., 3D printing) and platforms like Steam reduce piracy through convenience, not copyright enforcement.
    • [1:30:01-1:37:33] Juani and Kinsella reflect on persuading others, with Juani advocating for reform to shift the Overton window. They exchange book recommendations, emphasizing the cultural acceptance of copying in some traditions, and wrap up with mutual encouragement.
Links Referenced:
  • Juani’s essay: “Ideas for Free: A Case Against Intellectual Property” (not publicly linked in transcript).
  • Kinsella’s work: “Ideas Are Free: The Case Against Intellectual Property Law” (mentioned at 1:35:54).
  • Kinsella’s book: Against Intellectual Property (2001/2008, referenced at 53:42).
  • Book recommendations: The Way of Chuang Tzu by Thomas Merton (1:36:27); To Steal a Book Is an Elegant Offense (1:36:55).

Grok analysis of Núñez paper:

Thank you for highlighting Stephan Kinsella’s specific argument regarding patent trolls and “bad patents” versus “good patents” and practicing entities, as articulated in his article [“Patent Trolls Are Preferable to Practicing Entities”](https://c4sif.org/2012/05/patent-trolls-are-preferable-to-practicing-entities/). I’ll incorporate this perspective into the analysis of Juani Nunez’s treatment of patent trolls in “Ideas Are Free: A Case Against Intellectual Property,” focusing on whether Nunez demonstrates awareness of Kinsella’s view that the core issue with the patent system lies not with trolls (non-practicing entities, or NPEs) or “bad patents,” but with “good patents” held by practicing entities. I’ll also reassess how this impacts the comparison of their arguments and Nunez’s engagement with Kinsella’s broader critique.

### Kinsella’s Argument on Patent Trolls vs. Practicing Entities

In the referenced article, Kinsella argues:
– **Patent Trolls (NPEs) Are Not the Main Problem:** Patent trolls, or non-practicing entities, are firms that hold patents but do not produce products, instead profiting by licensing or suing for infringement. While often criticized for exploiting weak or overly broad patents, Kinsella contends they are a symptom, not the root issue.
– **“Good Patents” and Practicing Entities Are the Real Issue:** Practicing entities (e.g., Apple, Pfizer) that use patents to produce goods or services wield “good patents” (valid, well-crafted ones) to secure monopolies, block competitors, and inflate prices. These patents, often covering critical technologies or drugs, cause greater harm by restricting market access and innovation.
– **Systemic Flaw:** The patent system itself, by granting state-enforced monopolies, is the problem, not the type of patent or holder. Trolls highlight the system’s absurdity, but practicing entities’ monopolistic practices (e.g., pharmaceutical price gouging) have a more significant societal impact.

This view aligns with Kinsella’s broader abolitionist stance in *The Problem With Intellectual Property* and *Legal Foundations of a Free Society* (Ch. 14 and 15), where he argues that all patents, regardless of quality or holder, violate property rights by imposing nonconsensual restrictions on others’ use of their tangible resources.

### Nunez’s Treatment of Patent Trolls

In “Ideas Are Free,” Nunez addresses patent trolls in the section “Patent Trolls, Biopiracy and First Come, First Served”:
– **Criticism of Trolls:** He describes patent trolls as entities that “create patents for items or particular ideas, with the hope of being able to sell them for profit, or to sue anyone who use them,” citing cases like NTP vs. Research in Motion ($612.5M settlement) and Eolas vs. Microsoft ($520M overturned). He notes some patents are for “obvious” ideas, exacerbating the issue.
– **Examples:** Nunez highlights lawsuits over minor or intuitive features (e.g., Apple vs. Samsung over rounded corners, Lodsys vs. app developers for in-app purchases), framing trolls as exploiting the system to burden innovators.
– **Broader Context:** He ties trolling to the “first come, first served” nature of patents, which allows non-creators to claim monopolies, but doesn’t explicitly distinguish between NPEs and practicing entities.

Nunez’s critique of patents extends beyond trolls:
– **Pharmaceutical Monopolies:** He extensively discusses practicing entities like insulin manufacturers and AbbVie (Humira), criticizing their use of “patent thickets” and “evergreening” to maintain monopolies, leading to high prices and restricted access (“Monopolies” section).
– **Systemic Harm:** He argues that patents, particularly on essential goods, create artificial scarcity and cost lives (e.g., 10,193 daily COVID-19 deaths linked to vaccine patent barriers), implicating the system’s design rather than just its abusers.

### Does Nunez Show Awareness of Kinsella’s Point?

**Awareness of Trolls as Secondary:**
– **Partial Awareness:** Nunez’s focus on patent trolls is limited to their exploitative lawsuits, aligning with Kinsella’s view that trolls expose systemic flaws. However, he doesn’t explicitly state that trolls are less harmful than practicing entities or that “bad patents” are a distraction from the core issue. His examples (e.g., Lodsys, Eolas) target NPE behavior, suggesting he sees trolls as a significant problem, which slightly misaligns with Kinsella’s prioritization.
– **Broader Critique:** Nunez’s extensive discussion of practicing entities’ monopolies (e.g., insulin, Humira) demonstrates awareness of the harm caused by “good patents” held by producers. His examples of pharmaceutical companies using valid patents to block generics mirror Kinsella’s concern with practicing entities’ market distortions. This suggests Nunez implicitly recognizes the greater impact of practicing entities, even if he doesn’t frame it as Kinsella does.

**Engagement with “Good Patents” as the Core Issue:**
– **Alignment:** Nunez’s pharmaceutical examples (e.g., insulin’s patent thickets, Humira’s 200+ patents) focus on valid, enforceable patents used by practicing entities to maintain monopolies, aligning with Kinsella’s argument that “good patents” cause significant harm. His critique of evergreening and patent thickets shows understanding that well-crafted patents, not just vague ones, enable monopolistic abuse.
– **Lack of Explicit Distinction:** Nunez doesn’t directly contrast NPEs with practicing entities or label “good patents” as the primary issue. His discussion blends troll lawsuits with practicing entities’ monopolies, indicating a holistic critique of the patent system but not a precise adoption of Kinsella’s hierarchy of harm.

**Citation Context:**
– Nunez cites Kinsella’s *Against Intellectual Property*, indicating familiarity with his work, but doesn’t reference the specific article on patent trolls ([c4sif.org/2012/05/patent-trolls-are-preferable-to-practicing-entities/](https://c4sif.org/2012/05/patent-trolls-are-preferable-to-practicing-entities/)). His arguments about pharmaceutical monopolies suggest parallel thinking or indirect influence, but he doesn’t explicitly engage Kinsella’s nuanced view on trolls vs. practicing entities.

### Impact on Comparison of Arguments

**Revised Comparison:**
– **Shared Systemic Critique:** Both Nunez and Kinsella view the patent system as inherently flawed, granting monopolies that restrict property rights and harm society. Nunez’s focus on pharmaceutical monopolies complements Kinsella’s emphasis on practicing entities’ “good patents,” reinforcing their shared opposition to state-enforced IP.
– **Differing Emphases:** Nunez’s attention to patent trolls as a notable issue slightly diverges from Kinsella’s dismissal of trolls as secondary. However, Nunez’s broader critique of practicing entities’ monopolies aligns closely with Kinsella’s core concern, mitigating this divergence.
– **Theoretical vs. Practical:** Kinsella’s argument is grounded in libertarian property theory, framing all patents as nonconsensual servitudes. Nunez, while citing Kinsella, adopts a more practical, example-driven approach, emphasizing real-world harms (e.g., healthcare access) over theoretical precision.

**Nunez’s Reforms and Kinsella’s View:**
– Nunez’s patent reform proposals (shorter terms, royalty-based model, tangible invention limits) in “Ending the Patent Monarchy” address systemic issues, including those caused by practicing entities (e.g., reducing pharmaceutical monopolies’ duration). These align with Kinsella’s interim reforms (e.g., shorter terms in [“How to Improve Patent, Copyright, and Trademark Law”](https://c4sif.org/2011/02/how-to-improve-patent-copyright-and-trademark-law/)), though Kinsella might view them as insufficient without abolition.
– Nunez’s royalty model could mitigate practicing entities’ monopolistic control by allowing competition with payments, echoing Kinsella’s compulsory licensing idea. However, Kinsella might argue it still legitimizes state intervention, conflicting with his view that all patents, especially “good” ones, are unjust.

### Conclusion

Nunez demonstrates partial awareness of Kinsella’s argument that patent trolls and “bad patents” are less problematic than “good patents” held by practicing entities. His critique of pharmaceutical monopolies aligns with Kinsella’s focus on practicing entities’ harmful use of valid patents, showing implicit recognition of their greater impact. However, Nunez’s emphasis on troll lawsuits without explicitly downplaying their significance suggests he doesn’t fully adopt Kinsella’s hierarchy of harm. His reforms address issues caused by practicing entities, complementing Kinsella’s interim proposals, but lack direct engagement with the specific troll vs. practicing entity distinction in the referenced article. Overall, Nunez’s arguments reinforce Kinsella’s systemic critique, with minor differences in focus and framing.

PAPER:

Ideas Are Free: A Case Against Intellectual Property

By @witheredsummer

Intellectual property is, perhaps, one of those topics few people ever care to consider. Whenever one goes out into the hellhole that is the modern internet and looks at mainstream political and ideological discussions, the topics that are always brought up are economics, state intervention (or lack thereof), sociocultural policies, or private property, but it is very rare for anybody to ever bring up patents, copyright, trademarks and the like. Many will say to themselves: “well, it’s not like it’s too important to discuss”, while others will outright refuse any discussion on the topic, believing that intellectual property is something as sacred as the Quran is for Muslims, the Bible for Christians, or Das Kapital for Marxists.

This essay you will now read (I certainly hope) might be by many considered the delusional ranting of a madman, the dangerous thinking of a “revolutionary murderous communist”, or the naive armchair opinion of “lolbertarian corporate bootlicker”. Therefore, many will choose to not even read it, believing that there is no value in debating such an idea that’s either “unimportant”, or “fundamentally correct”, but those two are the notions I will proceed to tear apart in the following sections: the misconception that intellectual property law is not as important as other topics, and the doctrine that it is a necessity for any functioning society.

Before I begin, though, I feel the need to make a disclaimer: I am not a legal expert, I am not a lawyer, a historian, a political scientist, a sociologist, nor do I have any more authority on this topic than you, my dear reader. I am merely someone with a strong conviction who has taken some time to do some reading, and some time to do some thinking. You should know this, because this means that my opinion is not beyond reproach, but it also shows that it is not necessary to have a PhD in intellectual property law in order to form an informed opinion on the topic and feel capable of debating it with anybody else.

A Short Introduction

 

Normally, what most people would do at the beginning of such an essay would be to introduce people to the concepts about to be discussed. I’d much rather skip this and get straight to the ranting, but I’ll instead cite this definition:

 

Intellectual property rights are rights over ideal objects, which are recognized by the material medium on which they are embodied.¹

 

This one definition comes from a digital Spanish-translated copy of Stephan Kinsella’s Against Intellectual Property, which itself is quoted from Tom G. Palmer’s Are Patents and Copyrights Morally Justified? The Philosophy of Property Rights and Ideal Objects². These are two short works I heavily recommend reading if you’d like to get a much more nuanced understanding of not only what intellectual property (from now on called IP, for simplicity) is, but also of the libertarian philosophical arguments against it. Just don’t mix up Stephan Kinsella with the Irish economist Stephen Kinsella, those are two completely different people with different ideologies.

 

Now, I also believe explaining the (somewhat abridged) history of intellectual property laws is crucial for context, as it will help readers grasp the core arguments of this essay.

The (Mildly Abridged) History of Intellectual Property

While in modern times (disclaimer: as of May 16, 2025. If you’re reading this a century from now it might be different, please consult your nearest cybernetic overlord for updated information) we normally think of IP as springing up during the early industrial era, the truth is that this concept dates all the way back to the times of Ancient Greece and the Roman Republic. During classical antiquity and up to the early modern period, many systems existed which granted a limited amount of ownership over ideas to those who (claimed to have) invented them; in Sybaris, an Ancient Greek city, a law was passed around the 3rd century BCE, whereby whenever a “cook” invented a “new exquisite dish”, they were the only ones allowed to make it for a duration of a year; in other words, the person who came up with the new dish was given a year-long monopoly over it. This claim comes from contemporary historian Phylarchus, and was later cited by Athenaeus in Deipnosophistae³. The veracity of the claim, however, is dubious, since the only testimony comes from a secondary source and no legal records exist of such law, but this anecdote simply echoes the various forms of IP rights that might have existed, and likely did exist, during this historical period.

 

Five hundred kilometers northwest of Sybaris, we could find Rome during the times of the Empire, where Roman law also included notions of IP protection, which were, however, much more liberal. There was no equivalent to modern copyright, so while books were written, sold and bought, there was no ownership over their content, only over the physical items which hosted the writing (as in, the books themselves). However, Roman society did recognize moral ownership, which meant that respected authors had their works acknowledged and defended as theirs by their peers, and plagiarism was socially condemned, albeit not punished by law, unlike trade secrets, which were legally binding through contracts, somewhat similar to the modern day, but with less state intervention in the matter.

 

The first formal patent law was, arguably, conceived in the year of our lord 1474 in Venetia, Italy, through the Venetian Patent Statute. This statute granted exclusive rights to inventors for their innovations for a period of 10 years, and the principles in this document remain as the core principles behind modern IP law. 150 years later, in 1624, England’s Statute of Monopolies was created, which limited the Crown’s power to grant monopolies and promote patents for “new manufacturers”. The Copyright Act of 1710 in the United Kingdom was the first document for copyright law, granting authors control over their work for a period of 14 years and recognizing creative works as property (trust me, this is where it starts to go downhill). The US Constitution signed in 1789, in its Article 1, Section 8, gave Congress the power to grant patents and copyright, and the first patent law in the US was created through the Patent Act of 1790¹⁰.

 

Once the 19th century arrived, all hope was abandoned. The Paris Convention for the Protection of Industrial Property of 1883 established the first international IP treaty and introduced the principle of national treatment¹¹ (equal rights for both foreign and local inventors). The Berne Convention in 1886 standardized copyright law internationally¹² (only in its signatory nations). Naturally, after the first half of the 20th century globalization began to take place, which prompted the all-powerful, all-benevolent United Nations to create the World Intellectual Property Organization in 1967 to promote and harmonize IP laws worldwide (and ruin everyone’s fun). The TRIPS Agreement of 1994 forced the members of the World Trade Organization to adopt minimum IP protection across all forms of intellectual property, globally enforcing IP¹³. The final nail in the coffin came with the US’ Digital Millennium Copyright Act (DMCA), which, unlike the 1978 hit “Y.M.C.A.” by Village People, is not good, and clearly replaces a Y with a D. The DMCA introduced intellectual property into the digital world¹⁴, effectively turning the act of copying 1s and 0s in a particular order into a crime; but you wouldn’t steal a car, right?

 

Before moving onto the next section, as a final note for the (not that abridged) history of IP, movements like the GNU¹⁵ in the 1980s, and Creative Commons¹⁶ in 2001 represented a countertrend by providing creators the capacity to share their work through more permissive licenses, while not entirely waiving their rights. These licenses are just two of many others which fall into the category of Copyleft, to which an entire religion should be built around due to their martyrdom.

Why Not IP

Quite frankly, there are not a few, not dozens, not hundreds, but perhaps thousands of arguments to be made against the concept of intellectual property rights. These might be based on natural law, utilitarianism, materialism, or even theology, if you will. Compiling an entire list of arguments against (and, perhaps, in favor) of IP is a job I’ll leave to someone who is way angrier than me (and has more free time as well). Instead, I will explore several points which will provide you—my most beloved reader, who likely stumbled upon this essay by sheer chance—with an understanding at both a philosophical and a practical level of why intellectual property rights are both immoral and detrimental to society, and why they should ultimately be vanquished (some day, fingers crossed).

Property As a Tangible Asset

The most common definition of property in law is, simplifying:

 

The right to the possession, use, or disposal of something; ownership.¹⁷

 

This definition for property implies that something can be owned, used, and disposed of at one’s will. However, we have an issue when we try to translate this concept into the world of ideas: ideas are not a tangible thing, nor are they truly finite. Traditional property rights came into existence to manage rivalrous goods, items whose consumption by any one party prevents simultaneous consumption by another (e.g. if you eat a banana, I can’t eat that banana). Physical assets are inherently rivalrous due to their natural scarcity, so the existence of frameworks to allocate these assets is completely logical and necessary, as they resolve any potential conflict of interest between two or more entities over any given item.

 

On the other hand, ideas are fundamentally non-rivalrous, as they can be infinitely replicated; my knowing or using of an idea does not prevent anyone else from knowing or using the exact same idea simultaneously. Applying frameworks designed for rivalrous goods over non-rivalrous, abstract concepts, only produces artificial scarcity where none would naturally exist. Now, IP does not directly map to traditional property rights, instead it tries to incentivize creation by borrowing property concepts, as it follows the belief that if you give a monopoly over a creation to its inventor, they’ll be incentivized to innovate and create new things regularly in order to profit off their efforts. Nevertheless, this grafting of property concepts onto the realm of ideas remains problematic, precisely because ideas lack the inherent rivalry and scarcity that forms the basis of traditional property justification.

 

The concept of IP law also criminalizes parallel thought. It can inadvertently stifle independent creation and future innovation. Because ideas can be developed in parallel, and expression often follows logical or common pathways, IP regimes can and will penalize those who arrive at similar expressions, inventions or conclusions independently. This creates barriers to entry, favors incumbents, and discourages legitimate creative or scientific exploration for fear of infringing on broadly defined would-be rights. If you’ve ever had any idea for a story, a movie script, an invention, or any other thing that you could claim intellectual rights over, the chances that at least one of the billions of humans who have existed, currently exist, and will exist, has had or will have the same idea as you do, are likely closer to 100% than to 0%. Thoughts flow like the currents of a river, constantly, and each drop of water flows in the same direction, independently, yet alongside the rest, so it is inevitable that any idea will have, at the very least, one highly similar counterpart by another thinker, if not a carbon copy of it, but these independent creations often go unrecognized or unrealized, as the odds of such thinkers connecting to discover their shared genesis are incredibly slim. For more pseudo-profound nature analogies, refer to the Tao Te Ching and Zhuangzi.

 

All of this is merely an overview of the inherently flawed problem with IP law when analyzed through the lens of natural law. I insist that, if you want a much more detailed explanation of this, you read the aforementioned book by Stephan Kinsella, in which he explains many of these concepts in-depth in a way so beautiful it might make you cry, and then promptly become a political extremist.

Monopolies

During the (somewhat abridged) explanation of IP law’s history, you might have noticed that the word “monopoly” came up a number of times. Be not confused: monopolies are not a side effect of IP law, but the point of it. As mentioned earlier, the justification behind IP law is, generally, providing incentives to creation and innovation by granting creators a monopoly over their own invention, or in other words, giving them exclusivity to sell their invention or work, and granting them the legal capacity to punish those who infringe on their rights over said creation. While this may sound sensible on paper, it becomes closer to hellish when put into practice.

 

The monopolies which arise from, in particular, patent law, are, effectively, state-granted monopolies. Now, let’s be fair: there’s not much of a practical issue if I create a machine which cuts oranges in 3 pieces, patent it, and get a legal monopoly over it, because such an invention is hardly practical or useful, and arguably, a majority of inventions that get patented are similarly fringe in utility and scope, so there’s little space for serious competition there due to the niche market they may be aimed towards. However, these monopolies do become a rather big issue when these patents are granted over goods which become widely popular, useful or necessary, and there are some real-world examples of this:

 

The United States’ patent system interacts with its pharmaceutical industry in a truly egregious manner, and this becomes particularly evident in the case of insulin. To be fair, there are many factors at play in the broader context of the US’ pharmaceutical industry, but patents undeniably play a central role. The original insulin patent was filed in 1923 and was sold for just $1 (yes, $1 dollar, or $18.70 as of the time of writing this), as its inventors believed it should remain widely accessible. However, modern insulin is no longer the same product. Since the 1980s, companies have developed synthetic “human” insulin and insulin analogs, which are slightly modified versions designed to alter absorption or efficacy. Each of these modifications, even if almost negligible, be it a formula adjustment, a new delivery mechanism, or a packaging innovation, has been grounds for filing new patents. This tactic is known as evergreening, and it leads to the creation of patent thickets: dense webs of overlapping intellectual property claims that make it nearly impossible for generics or biosimilars to enter the market.

As a result, major pharmaceutical firms get to keep long-term monopolies over various insulin formulas. These monopolies are then aggressively defended, often through litigation, against any organization attempting to introduce affordable generics, and because insulin is a life-sustaining drug with highly inelastic demand, companies can set exorbitant prices with little fear of losing market share. In most cases, these inflated costs are absorbed not directly by patients but by insurance providers or the state, allowing the price bubble to persist. This creates a perverse market dynamic whereby the demand doesn’t decline in the face of higher prices, and the intermediaries shielding patients from direct costs further dull the pressure to reduce prices, thus sustaining a constant loop of, in simple terms, regulated market exploitation.

There are many other such examples within the pharmaceutical industry only. To name a few:

 

  • EpiPen, a drug to treat serious allergic reactions, albeit cheap and generic, was historically highly priced (up to $600) in the United States, even when its cost of production was around $10¹⁸, because only one company owned the patent to the auto-injector delivery system it uses. Its price was ultimately capped in 2025.
  • Truvada is one of many HIV antiretrovirals, and it was developed, ironically, using taxpayer-funded research, yet it was priced at over $2000 in the US¹⁹ until generics entered the market, and its price dropped to about $18²⁰.
  • Humira, used for various autoimmune conditions, had over 200 patents filed just to delay competition and had the entry of any biosimilars to the US delayed until 2023, before which its patent owners made 20 billion USD a year in profits, making it likely the most lucrative pharmaceutical monopoly in history²¹.
  • Cochlear Implants, which are life-changing implants for people with severe hearing loss, have multiple very strong patents assigned to Cochlear Limited on both their hardware and software²², so these devices can cost tens of thousands of dollars and only a few competitors exist in the market.

 

Beyond that, there are dozens of other products worth mentioning across all sectors of the economy. For instance, patent laws on genetically-modified seeds (largely owned by companies like Monsanto and Bayer) ban farmers from reusing seeds from patented crops and legal actions can be taken on farmers who do (see Bowman v. Monsanto Co.²³), even if it was by accidental contamination. Many Lithium-Ion Batteries are unique and patented, with proprietary designs not shared between different manufacturers, which has negative effects when replacements of batteries for existing products are necessary.

 

I could certainly spend about 160 pages of this essay mentioning IP monopolies and how each one of them has negative consequences on consumers, but I imagine that, by now, you get the idea. These monopolies do not only hurt people who want to buy a new house appliance or a snack and have few alternatives to choose from, but they literally create artificial bubbles which delay the progress of crucial fields of medicine and research by prohibiting competition, and through this, massively hurt the people whose lives depend on many of these life-changing technologies, who end up unable to access them due to artificially-inflated prices.

 

To put this into numbers, it is estimated that an average of 10,193 people died from COVID-19 each day that the British government had blocked attempts to waive intellectual property for coronavirus vaccines and treatments²⁴. In the early 2000s, the cost of antiretroviral therapy for HIV/AIDS was between $10000 and $12000 per year²⁵, which made treatment for patients in low-income countries almost completely impossible, which most certainly led to the deaths of thousands of people from a disease which had become easily treatable. Similarly, about 80% of deaths due to diabetes, cancer and cardiovascular conditions happen in middle and low-income countries²⁶, with many of these deaths, though preventable with access to innovative treatments and drugs, occur because international IP law limits such access, and this pattern also extends to malaria, tuberculosis and other ailments²⁷ where access to cures or effective treatments is hampered, in part, by restrictive patent laws. In simpler terms, patent laws cost lives²⁸.

Not-So-Free Sharing of Information

Another issue of great weight comes in the form of copyright. If patents limit people’s access to physical items, copyright limits people’s access to ideas and knowledge. In the modern day, the vast majority of academic research, scientific books, non-fiction works, statistics, educational books and most other forms of concrete data capable of providing progress are locked behind copyright law. This means that vital information that can lead to the faster development of new technologies and which may allow people to further their skill sets and technical expertise, thus allowing further personal, and by extension, societal growth, is protected by, in simple terms, paywalls.

 

Now, there’s more of a point to be made with copyright than patent law: creative and research works require much more personal thought and unique creativity than the creation of copious physical items which may (or may not) serve a practical purpose. There is a lot of mental effort and time which goes into writing a novel, researching data for an academic paper, or creating an entire movie or videogame, and one of the main incentives to do this might just possibly be the prospect of profiting off these works. Nevertheless, there reaches a point in which one has to put into doubt whether forcing everyone to pay a fee just to be able to read a particular text, without allowing them to even replicate it, is not outright harmful to society at large.

 

What copyright does is, practically, limiting the free flow of knowledge and information. This outright limits education and progress in all fields: medical, economic, technological, artistic, academic, agricultural, cultural and pretty much any other field that ends in “ic” or “al”. Data which might be useful for better economic calculations or business management might require huge fees to be accessed, or crucial findings in medical fields like neuroscience might be only available in a paper that costs hundreds of dollars to just read and download as a PDF. While it can be argued that most of those with the capacity to put that knowledge into use tend to own or belong to organizations with considerable capital, there’s still a considerable portion of the population, be it individuals, for-profit or non-profit organizations, which could use that data to further develop concrete goods which could, in some cases, lead to the improvement of many lives. What this inevitably does is turn knowledge into a commodity instead of an accessible, shareable public good.

 

Inevitably, I must once again make a pledge for the poor and remind the reader that, at the end of the day, copyright, just as patents, hurts those with the least resources the most: as someone who grew up in a lower-income household, I went to a lower-income school, in a lower-income city, in a middle-to-lower-income country. In school, we used various books, most of which we had to photocopy, since purchasing them was beyond the budget of most of the students in the school, myself included. This, in practice, meant that in order to get education, both us, the students, and our teachers had to resort to criminal behavior; according to both national and international law, photocopying the books we used was a violation of copyright law, and was therefore punishable. Of course, nobody was going to enforce it, but as we move towards a much more technology-reliant future, new protections against copying and sharing copyrighted works are introduced, and we might not be far from a future in which technology prevents someone from making a photocopy of a copyright-protected work. My situation is not unique, but widespread in most of the developing and underdeveloped world, and the education of billions requires breaking the law; once we get to a point in which this becomes punishable, or impossible thanks to technological advancements on IP enforcement, we’ll be facing a world where education becomes, once again, inaccessible for those who cannot pay for it.

 

That obviously doesn’t just happen with education, but with all other areas of progress for the people of non-developed nations. Just like copyright applies to written works and prevents university students from accessing academic research or relevant books (such as in my case with most computer science and programming books I used during my higher education), or various companies from accessing market data or useful technologies, copyright also protects software, which, in the modern day, has become a vital part of both economic output and people’s daily lives. This results in many companies in underdeveloped nations, particularly smaller ones, having to use pirated operative systems and pirated software in order to carry out their economic activities, and while free alternatives exist for most popular software, the truth is that many have tools that are not available elsewhere (sometimes also due to, you guessed it, patenting). It is not hyperbolic to claim that if you go anywhere outside of OECD nations, you’ll find that a majority of local companies are running pirated versions of Microsoft Office, Windows 10/11, Adobe Photoshop, or any other software commonly used²⁹. This, again, creates a liability for all of these companies, who can be sued by the proprietary owners of the software they are pirating, and as new DRM technologies are developed, it becomes increasingly harder to pirate these necessary programs, which in many cases are simply unaffordable since few companies offer regionalized pricing for their digital products, which is then worsened even more by the fact that many software characteristics, even the DRM technology itself, end up protected by patent law, so, for instance, small software development companies from poorer nations might not even be able to afford the methods that will prevent their own software from being pirated by, ironically, people in high-income nations who can actually afford them.³⁰

Erasure of History Through Copyright

I could talk a lot more now of how copyright on software also hurts education, or about how that same copyright limits people’s access to entertainment and creative works, but you can already get an impression of this without having to be told about it, so what I’d like to focus on now is somewhat more niche in scope, but extremely important nevertheless, and it is about how copyright creates artificial scarcity and can, adversely, erase entire works from existence.

 

You have probably heard of lost media; books, videos, videogames, movies, songs and other works that are no longer accessible because they’re lost to time and hard drive failures (requiescat in pace, MySpace). While most of these are lost due to copies not being saved, or the few that did exist being destroyed, we are currently looking down the barrel of a world where, even if you can store anything in thousands of computers at once, media will begin to be erased from history, as if it never existed, because of copyright preventing people from archiving it or sharing it. In recent years, the Internet Archive, the largest digital archive for all types of media throughout history, has been the target of multiple lawsuits which aim to prevent the organization from storing books³¹ and music³² on their servers, effectively forcing them to possibly have to erase these files, many of which are digitalizations of highly rare books, records which can no longer be found anywhere, and many other types of media protected by some sort of IP. These lawsuits might just be the beginning of a long list of future legal issues, since the Internet Archive stores millions of works protected by copyright, and at any given time their owners may come knocking. In the worst-case scenario, the archive could be forced to shut down, not only erasing millions of works (with tens of thousands of them nowhere else to be found), but also erasing over 25 years of the internet’s own history. The loss of knowledge from such an event would be the equivalent to the Library of Alexandria burning 2 million times over. Copyright seeks to protect the works it reigns over, yet, it’s capable of erasing them from existence entirely just to ensure that it doesn’t get undermined out of fear of a slippery slope.

 

Another issue related to copyright is that it renders a lot of media simply inaccessible, even if they’re not necessarily “lost”. For instance, let’s use a recent example: The Crew was an online, open-world racing videogame released in 2014 by Ubisoft. In December 2023, the game’s servers were shut down, and the game not only delisted from digital stores, but erased from people’s accounts; those who had bought it no longer had access to it, with the justification used by Ubisoft being that “they paid for a license to play the game, not for ownership”. This led to widespread criticism³³, a lawsuit³⁴, and made a dedicated community of its players start working on a fix to allow players to play the game again, but only through piracy, since it is no longer legally accessible. This is not the first time this happens to a videogame; the same happened years earlier with Battlefield 2, Battleborn, Darkspore and hundreds of other games³⁵ which have been rendered completely inaccessible even for those who paid for them. Similar things have happened with movies (such as The Flamingo Kid), music albums, books and other media, all of which simply couldn’t be saved because of copyright-related lawsuits.

 

There are, also, thousands of works out there for which few remaining copies exist, so the few people who do own those copies and are willing to sell them generally put high prices, and creating copies of that media is, you’re right, illegal, because they are still protected by copyright, in some cases even after their original owners have passed. For instance, while not at all scarce, the music of The Beatles, although over 60 years old now, is still protected by some of the most aggressive copyright and phonographic rights we have ever seen, even though half of the band’s members are now writing songs beyond the wall of eternal slumber, as well as many of those who were involved in the recording and production of those albums. Copyright ends up forcing people who want to experience certain works to have to do it unlawfully, because there simply exist no legal means to do so, even if there, theoretically, should be.

Artistic Erasure and Replacement

Not only does copyright erase history, but it also is capable of replacing or erasing culture and art. These days, few companies own entire catalogs of movies, music and books, and willingly decide to vault them or limit access to them. The reasons are many, but in general, they’re related to keeping the company’s name and catalog relevant and associated with a certain type of media or crowd. This effectively works to homogenize culture.

 

For instance, much of the pop music you may listen to on most radio stations, whether you’re American, German, Brazilian or Papuan, is from America or the Anglo-speaking world, and this is mostly because said music is backed by large copyright-holding companies, which have the necessary capital and influence to push this music into the mainstream through advertisement in streaming platforms, TV, social media and through massive tours for their artists. This causes local music in non-English languages, and especially unique local genres, to slowly get erased, since the general crowd begins moving towards music in English that is radio-friendly; this is a huge reason why, in the modern day, even most internationally-renowned artists from non-English-speaking nations sing in English. It’s because English music sells because labels monopolize it and push back other music. The copyright structures that these companies use clearly give an incentive to singing in English, which in turn makes younger generations grow up experiencing a completely different type of culture which replaces their own. What many of these labels also tend to do is vault the works that can intervene in the process of promoting their most accessible music, and through the purchase of smaller copyright-holding companies which own the rights over various songs by artists of different nations, they can effectively take them out of the public light and punish anyone who tries to bring them out again.

 

Disney and Hollywood are two of the main culprits behind worldwide cultural homogenization. Disney has popularized and copyrighted traditional folk stories and fairy tales, like Snow White, Cinderella or The Little Mermaid, altering them and using their copyright monopoly to pass them as the “authentic” or “definitive” versions. This also works as a form of promoting cultural or historical erasure, as they did, for instance, with the 2025 Snow White movie³⁶, where Snow White was played by, ironically, an olive-skinned actress, and in which the movie erased much of its original lore (for instance, the Prince is no longer an actual prince, but a Robin Hood-style bandit). This is an attempt to supersede the original story, sanitizing it for modern audiences while leaving behind its original intentions and messages. Such traditional stories are often in the public domain, but once modified and shielded by armor-clad copyright by big players, the original tales are often suppressed and erased from collective memory. Only a handful of examples avoid this fate, one of them being Anne of the Green Gables, which although in the public domain, is indefinitely trademarked, and the descendants of the original author only allow faithful recreations of the story in other forms of media.

 

As was mentioned at the beginning of the previous paragraph, Hollywood films are also plagued by similar problems to those in Disney movies: they often readapt stories from other media or earlier productions, heavily modifying them, to then export them abroad, where they’ll be consumed far more than local entertainment since, once again, there’s much more profit to be made through Hollywood’s complex and huge copyright structure, making local adaptations of these works practically impossible due to extreme licensing costs, which has practically crippled the film industries of most other nations, pushing them into a niche while audiences flock to American productions instead.

 

This all also adversely affects remix cultures and fan works (such as Japanese dōjin culture). Making remixes of existing music has become increasingly harder, as DMCA strikes tend to target any copyrighted music uploaded to mainstream platforms (even in some cases where a recording is quoted for just a few seconds, or even when it’s played by its own author³⁷). Similarly, transformative works, or those which use characters or concepts belonging to copyrighted franchises (like Harry Potter or Star Wars) tend to be targeted if any profit is made out of them (directly or indirectly). This can happen even with fan-made artworks or memes.

 

Perhaps most importantly, when copyright over works is held not by their authors, but holding companies, this creates the potential for widespread censorship. Let’s imagine a scenario in which, for instance, the music label Epic Records has rights over the entire discography of a certain artist, but this artist then says something highly controversial in public. While (currently) far-fetched, now the company has the possibility to censor the entire catalog of said musician³⁸. If we currently look at the increasingly strict regulations on speech across the world, in particular in the European Union and the United Kingdom, we might not be very far from having governments pressure copyright holders into censoring political opponents, in what would be the equivalent to modern book-burning.

Patent Trolls, Biopiracy and First Come, First Served

Imagine you implement some niche technology to an invention of yours, simply because it makes said invention look better, or gives it a small quality-of-life improvement, then you decide to market it, and before you can say “Man, I hate patent laws”, you’re facing a multi-million-dollar lawsuit because that tiny tweak you made was already patented about a decade ago by some small company dedicated to making the lives of people like you impossible. You just fell victim to a patent troll: Patent trolls are basically people who create patents for items or particular ideas, with the hope of being able to sell them for profit, or to sue anyone who use them. In some cases, the ideas being patented are obvious, which makes this even more ridiculous. Many examples of this exist throughout history, so to give some examples:

 

  • NTP, Inc. is a holding company which, in 2001, sued Research in Motion (the creators of BlackBerry devices) for infringing on its patent related to wireless email technology³⁹. In 2006, RIM had to settle for $612.5 million to avoid BlackBerry services from being shut in the United States⁴⁰.
  • Eolas Technologies sued Microsoft in 1999, claiming that Microsoft’s use of plugins in their Internet Explorer browser infringed on Eolas’ patent for embedding interactive elements in a web page. Microsoft almost had to pay $520 million in damages⁴¹ before the case was overturned on appeal, but the legal fees were nevertheless lost.
  • Lodsys sued many small app developers in 2011 for allegedly infringing on its patents by using in-app purchasing mechanisms⁴². Lodsys did not create any apps or technology but sought licensing fees from developers. Many developers were forced to settle or pay licensing fees because they couldn’t afford the legal costs of fighting the lawsuit, even though some argued that their use of the technology was covered under other licenses.
  • Apple sued Samsung in 2011 over infringements of various design and utility patents, including those covering rounded corners on devices and the “slide to unlock” feature⁴³. Apple was originally to be awarded over $1 billion dollars in legal damages by the lawsuit⁴⁴, but the legal battle was eventually settled for around $500 million⁴⁵.
  • Intellectual Ventures is a known patent troll holding company which has sued many others, such as Google, Microsoft and Apple, over infringements on many simple or obvious features included in their products. Most of those lawsuits were settled outside of court.

 

These are just some cases, but hundreds exist throughout history. This comes to show that patents can and tend to work as liabilities for companies, big and small, since others can simply force them to pay for using certain features or ideas in their products, even though they might be ridiculously common or intuitive.

 

Another issue with IP law is that, in many cases, it works as a first come, first served system, commonly seen through biopiracy. What this means is that whoever registers a certain patent or work first owns the monopoly on it, even if they were not its creator. While there exists protocols and laws that try to avoid this from happening, it has nevertheless happened many times, and these protections usually don’t apply over communal ownership very well. This leads to certain techniques, inventions and even creative works being taken ownership of by people who had no interference in their discovery or creation. Since I really like bulleted lists, examples include:

 

  • In the 1990s, the US Department of Agriculture and W.R. Grace & Co. patented a method of extracting neem oil from the Neem tree⁴⁷. The Neem tree is native to India and has been used for millennia for traditional medicine and agriculture. Since this was widespread generational knowledge, no protection existed for it, and the Indian government had to wage a lengthy battle for the patent to be revoked in 2000, at least in the European Union⁴⁸.
  • In 1997, the American company RiceTec was granted a patent on “basmati rice lines and grains⁴⁹. Basmati rice is a traditional agricultural product native to India and Pakistan. The patent claimed innovations that were already part of the traditional practices of rice cultivation in those regions, and it was eventually mostly revoked as it was considered a form of biopiracy⁵⁰.
  • Another case pertaining to India is the Turmeric patent which was granted in 1995 to American researchers on the use of turmeric for healing wounds⁵¹, which, yes, was already known in India for centuries. Yes, the patent was revoked in 1997 too⁵².
  • Ayahuasca is a traditional spiritual medicine used by indigenous groups in the Amazon basin. In 1986, an American named Loren Miller was granted a patent on a specific variety of the plant⁵³, and of course, the patent was protested by Indigenous organizations of the Amazon, ultimately considered biopiracy, and revoked in 1999⁵⁴.

 

Yet another pressing matter with this is the fact that legal battles over patents and copyright are expensive. Expensive generally means that those who have fewer resources have fewer chances of winning any legal battle, which is exactly the case with the next examples (I love bulleted lists!):

 

  • Robert Kearns was an American inventor who invented and patented the Intermittent Windshield Wiper. Despite the patent, companies such as Ford and Chrysler began using the technology without his permission. Kearns did win the lawsuits he filed against these companies, but it came as more of a pyrrhic victory for him, since it took years and large sums of money to cover legal fees⁵⁵.
  • The Wright Brothers, credited for inventing the first flying airplane, had also patented the design of the aircraft and its innovations. Glenn Curtiss, another pioneer, began using similar technology in his own aircraft, which led to extensive legal battles with the Wright Brothers, who were much less financially capable of seeing them through⁵⁶.
  • The Happy Birthday song was originally written by the Hill sisters in the late 19th century. However, Warner/Chappell Music claimed ownership of it and collected royalties for its use. This copyright was only revoked in 2015 after a lawsuit was filed in 2013⁵⁷, even though the song should have been public domain. For decades, small companies, filmmakers in particular, were forced to pay royalties for featuring the song.
  • Edwin Howard Armstrong is credited with inventing FM radio technology in the 1930s. The well-established AM industry opposed his creation, and RCA copied and used it without proper licensing, which led Armstrong into an extensive and expensive legal battle with not only RCA, but many other companies which used his patent without permission⁵⁸. These prolonged battles took a severe toll on his finances and mental health, leading to a breakdown and paranoia. The immense strain also deeply affected his family, leading his wife to be committed to a mental hospital after she reportedly jumped into the East River, and tragically, leading to Armstrong taking his own life at age 64⁵⁹.

The PPP Doctrine: Patent, Protect, Perpetuate

A final issue that I would like to bring up, out of so many I could continue to mention, is how IP laws create their very own need to exist, as a self-reinforcing legal construct. The very existence of laws that protect intellectual property creates incentives which justify its own continuation. It’s the prisoner’s dilemma applied to intellectual property law⁶⁰; if nobody patented their creations, then ideas could freely flow and build upon each other, like it happens in the real world without need for IP, but because IP exists, anyone can patent something and restrict its use, which then forces creators to patent their ideas defensively: not because they want to restrict others, but because they don’t want see themselves restricted from using and marketing their own idea. It’s a legal arms race.

 

Some will argue that there’s an inherently important aspect to IP, and it is that it incentivizes innovation, therefore, IP exists justifiably as a mechanism of incentives to prevent technological stagnation. Ironically, this is exactly what IP promotes: conformity in exchange for profits. Once you own a monopoly over a product, it’ll take someone else to create a similar product that’s better and/or cheaper, and which does not violate your patent for you to actually have an incentive to improve upon your invention or invent something new. IP only manages to create cycles of progress followed by a long stagnation.

 

What is perhaps most absurd when it comes to the justification of IP as necessary for progress is the claim that its existence is what drives innovation. Now, I don’t wish to insult the intelligence of any one of my cherished readers, but I must admit that certain followers of a female Russian-American author I will not name are, just perhaps, unaware of human history prior to the Industrial Revolution. Many like to claim that IP law has direct causality with the accelerated technological progress humanity began experiencing during the late 18th century, but this is a clear case of correlation; the popularization of IP law merely happened to coincide with the Industrial Revolution, which brought with it massive technological and scientific advancement which created incremental loops of further development. Even some of the most revolutionary inventions of the time, such as the early versions of the steam engine, were not truly patented⁶¹. This argument falls apart even further when we consider that even when some previous IP laws existed (such as the Venetian Statute), they didn’t particularly have any noticeable effect in innovation, otherwise, Venetia would have become the greatest technological hub of the early modern period, which it was far from, as, in fact, the city came to be in decay by the 17th century.

 

The argument in favor of IP as a must for innovation is fundamentally flawed, not only through practical examples, but through sheer praxeology; if people decided to innovate seeking to monopolize their inventions, then, logically, you wouldn’t be reading this, in fact, you’d probably be eating poisonous mushrooms or banging rocks inside a cave, since no human would have even cared to move past paleolithic technology, as no incentive would have existed to create new advancements which everyone else could copy. Unlike what defenders of IP like to believe, it is evident that IP laws exist because of emerging innovation cultures throughout history, rather than the other way around, and that IP is better understood as a political and economic tool to consolidate or protect gains, rather than a system of incentives for entrepreneurship and creativity.

IP in the AI Era, or the Revelation of a Contradiction

Originally, I hadn’t thought about artificial intelligence (AI) and its relation to intellectual property while I was writing this essay; it came to me while I was trying to fall asleep a few days later, and ever since, I’ve been thinking and analyzing this relationship, which only led me to develop an even greater hatred for intellectual property laws, which is only rivaled by Ted Kaczynski’s hatred of modernity (minus the mailing people explosives part and probably soon to include the going-into-complete-reclusion part). The section you’re about to read is considerably extensive relative to the previous ones, and I deem it to be a sub-essay of this entire manifesto.

 

AI is not a new concept; humans have fantasized with machines having human-like intelligence for centuries, and the first incursions into the field of artificial intelligence began taking place way back in the 1950s, by the likes of Alan Turing and John McCarthy. However, it took decades and many “AI winters” for us to reach the modern day, where a computer can use complex probabilistic algorithms to recognize speech, generate expressive human-like voices, generate realistic videos and images, or write entire essays on any given topic (worry not, fellow humans, I can assure you this essay was written by another Homo sapiens). This huge development in AI is the result of investment and further technological advancement in the fields of neural networks, deep learning and computational complexity, and is soon to begin being hindered by… do I have to mention it?… IP, it’s IP.

 

For an AI to be able to recognize a face or a word, or to be able to generate any type of content, it has to be trained on preexisting data. Different models are trained with different data based on their scope, be that human faces, texts, speech, or a combination of these and other things. While there are hundreds of different implementations of AI which aid people, including us, on a daily basis, perhaps the most amazing, most useful, and most mainstream ones are generative AI models like GPT, Gemini or Stable Diffusion; we use these to ask questions, make images, learn about new things, get help with work or our studies, or just to have some fun. These models are capable of such massive and varied output thanks to them being trained on a massive amount of data from thousands of sources, upon which are inevitably included copyrighted works.

 

As of the time of writing this, the world of AI is pretty much the 21st century Wild West: mostly lawless, primarily regulated through common interest and professional standards, and sitting there in its relative freedom waiting for regulators to get their hands on it. Because of this, we’ve seen some extremely massive leaps in the quality and capacity of AI models just every a few months: we went from having videos of a deformed Will Smith eating what seem like noodles to videos that are almost indistinguishable from reality, as we went from chatbots which struggled to do anything more than basic reasoning to building entire small videogames using a few prompts. I believe this to be greatly thanks to the fact that the different companies investing in the field have been left to their own devices, not having anyone putting a gun on the table and forcing them to do or not do something. If we look back in time, we can see how other fields, such as aviation, genetic engineering, pharmaceuticals, and modern telecommunications (as well as the internet itself) saw a golden age which was eventually halted when oversight by state agents was imposed upon them. While sure, we can point out some reasons for that: early aviation was dangerous, genetic engineering risked having us playing God the wrong way, pharmaceuticals led to such things as the thalidomide disaster, and the early internet was a land of near-null data privacy and unrestricted speech. However, what all these fields share is that after regulations kicked in, they saw themselves crippled, to different extents, and for the most part, the consumer paid the price, but most importantly, it equipped the State itself with the capacity to monopolize advancements in these fields for its own nefarious purposes (firebombing of civilian settlements, drone warfare, the Tuskegee Syphilis studies, weaponization of Anthrax, mass surveillance, and the list goes on and on).

 

Right now, we are in a rollercoaster climbing up high, waiting for its carts to fall down the tracks into the arms of bureaucracy, and, inevitably, the expansion of IP law to completely encompass the entirety of the AI field and all of its byproducts. As I mentioned earlier, AI models need to be trained on preexisting data, and what better data to train an AI like GPT than thousands of books, videos, music, research papers, images and such? Surely nobody would care to put their o-so-righteous intellectual property rights over the advancement of what is perhaps the most revolutionary invention in human history… right…? Who are we kidding? Of course, they would! We are already seeing the first legal battles fought by IP owners against AI tech companies over infringement of their would-be-rights, either because they used them to train their models and/or their models generated content similar to their intellectual works. For one last bulleted list from yours truly:

  • Authors Guild v. OpenAI: Various authors, George R.R. Martin and John Grisham among them, argued that OpenAI used pirated or unlicensed datasets that included their entire works to train their models, and that said models can summarize, emulate or generate content based on their writings⁶².
  • Silverman v. OpenAI / Meta: Sarah Silverman, an author, joined others in alleging that OpenAI and Meta used her copyrighted books without permission during AI training, including the claim that the copies used were pulled from shadow libraries like LibGen or Sci-Hub, and that the models by these two companies could generate content very similar to their own works⁶³ ⁶⁴.
  • Andersen v. Stability AI / Midjourney / DeviantArt: Three visual artists sued Stability AI (developers of Stable Diffusion), Midjourney and DeviantArt, claiming that their visual styles and copyrighted works were used to train AI art generators without consent, which in turn allowed people to generate art similar to theirs, undermining their work and diluting the market⁶⁵.

 

In the near future, we should expect to see countless similar lawsuits from authors, artists, musicians, public figures and others, claiming that any given company violated their intellectual property rights. It is very clear that the arrival of the AI era created a huge gray zone for copyright law, and while some might think that this is due to the state not having yet intervened to set the rules of the game, I propose a deeper, more unsettling explanation: AI has exposed the fundamental irrationality of IP law by removing the human element to the act of creating things, and therefore, it has effectively nullified the idea that IP has to exist in order to reward creative thinking and intellectual effort. What we all knew, but refused to acknowledge, was revealed by this new technological revolution: that the process of creation, at its core, is iterative, derivative and procedural, the very same traits that IP law claims to elevate above mere reproduction.

 

This long-delayed revelation inevitably forces those who believe in the legitimacy of intellectual property to face that AI is not a disruptor of IP, but rather a mirror of its inherent contradictions, and it puts under the light what most of them would rather blissfully ignore: that this system was never designed to protect culture or creativity, but rather to control and enclose it. Now that computers are capable of generating content not unlike a human, the defense of IP becomes more about gatekeeping access to cultural and intellectual capital, in an effort to assert human exclusivity, not on moral grounds, but on economic ones, and in the process, slowing down progress in exchange for prolonged profiteering through State-granted monopolies.

 

Historically, intellectual property law has relied on the romantic notion of the solitary genius, of the individual who conjures originality ex nihilo, as if by divine inspiration. AI has decisively undermined this fantasy by producing seemingly creative outputs while lacking a human brain, and by extension, the ability to reason, feel or decide. AI is capable of creating not because it is inspired, but because it has access to large amounts of informational input, and the capacity for pattern recognition and recombination, which, if we stop to think about it, isn’t too different to how we ourselves create things. Where does a musician get the inspiration to write music? Normally, from other musicians; someone who has never listened to music can’t be expected to be able to write a song. Where does an artist get the inspiration to make a painting? From the works of art of both nature and men; someone who spent his entire life inside an empty room wouldn’t be capable of painting anything more than abstract nonsense. Where did I get the inspiration to write this essay? From reading various authors and the opinions and arguments of my fellow ideologues as well as those of my opponents; had I never cared to read or debate on the topic of IP, then you wouldn’t be reading this right now. This all brings us back to a core philosophical inconsistency of IP: how can we possibly claim ownership over an idea or creation when we had to draw inspiration from others’ works? Why shouldn’t those people have ownership too over what we make, considering they were instrumental for our own creativity?

 

At this point, I can imagine my many detractors might not have been convinced by my appeal to rationality, but with any luck, a more practical, consequentialist approach might sway their opinion. Maybe you, the reader, are not only a detractor, but in fact, you are someone wearing a fancy suit, in some fancy office, living off the money of taxpayers like me, thinking of how great it would be to regulate the entirety of the AI field just to spite me for writing this essay, and while this might sound like a paranoid delusion, this imaginary bureaucrat is the example of a person whose job would be to halt human progress. Such a bureaucrat likely exists, and like many others, is currently sitting somewhere, thinking of some sort of new arbitrary project to propose to their legislature so that the development of AI can be regulated and slowed down over ethical concerns such as “what if AI is used to spread misinformation?” or “what if AI is used to promote or teach violence?”. If I had to ask a couple of questions to said politician, they’d probably be “what if slowing down the progress of AI prevents us from fixing climate change or curing cancer?” and “what if AI is used by the government itself to spread misinformation and teach violence?”. Thanks to AI, we are now able to make huge progress in vital fields for human survival and progress: we can find better ways to treat illnesses, find new cures, predict economic trends, find new ways to fight issues like climate change, and countless other things. How could we possibly trust the State to regulate a technology that so far has been extremely beneficial, on ethical grounds, knowing that the State itself is the main perpetrator of unethical behavior in the world? Are we willing to give the bureaucrats the power to turn a revolutionary technology into a tool to wage war against its own population and other states, all through “smart regulations”, like they did with drones, biotechnology, drugs, encryption and currency itself?

 

Perhaps what is most concerning about the prospect of regulating AI and its associated fields is that there’s no objective line to be drawn. Who determines when AI infringes on copyright law and how? At what point does IP over a creation become void if it was made with AI assistance? Who even is at fault? Imagine you ask a large language model to generate a particular story about a young guy with glasses who goes to a school for magicians, and then find himself trapped in a battle royale deathmatch with other students, and at some point one of the characters also happens to explain to the protagonist the intricacies of wealth and income inequality in the United States and Europe since the 18th century. Then, you share that story publicly. The AI might be infringing on the copyright of multiple different authors by creating that story. You would probably be trespassing on their copyright as well for sharing it or profiting from it. The question, though, is “who’s really at fault here?”. The owner of the model you used? All they did was probably train the AI on large datasets which might or might not have included the works of those authors, legally or illegally, but the AI did not generate that story by itself; it required you to engineer a specific prompt to do it, so blaming the company for allowing a user to use the tool it offers in a way that violates IP wouldn’t be different to Universal suing Gibson because someone recorded and sold an altered cover of Helter Skelter by The Beatles as a derivative work using one of their guitars to record it. Maybe you are at fault, but you did not write a single word of that story, all you did was share it, perhaps even through the AI application’s own system (such as GPT’s sharing feature), but if you get sued, then what about every single person who used AI to aid in their creative writing process? I certainly doubt anyone has the time or money to sue millions of individuals. Don’t forget that in this hypothetical scenario, you shared or profited from the AI-generated story, but the lawsuits I previously mentioned only required the AI model to generate text similar to those copyrighted, without the need for any user to share it. In that case, why don’t authors sue others for mimicking their style or concepts? Why don’t artists file lawsuits against people who replicate their paintings for practice, or musicians drag other musicians who recreate their music as tributes or draw inspiration from them to court over infringement of their intellectual property rights? The reason is simple: we culturally and legally accept that learning, imitating and getting inspired by the works of others are essential parts of the creative process. If it’s okay when we do it, then why is it suddenly theft when a machine does it at scale and without any trace of the human ego?

 

So, how do we even regulate that? Do we ban companies from training their AI with datasets containing copyrighted works? Make them have to filter through millions of texts, images, videos, internet comments and more just to make sure none of it is copyrighted? That doesn’t at all prevent their AIs from generating a string of alphanumeric characters which might resemble or be the exact same as something that’s been copyrighted, after all, an AI is essentially not different to a bunch of monkeys banging on keyboards in the way it works. So, maybe we allow usage of copyright if the IP holders allow it? Surely it’d be a quick and easy process to ask hundreds of thousands of individuals and organizations permission or licenses to fairly use their works, with said licenses often expiring after a certain amount of time, not to mention it is realistically impossible to regulate the way people will use the AIs trained with that data as to ensure the models will not in any way infringe on the licensing conditions. There is simply no way we can create an objective set of rules determining what data AI can be trained with, and what it can or can’t generate, because AI is practically an amalgamation of the entirety of human creativity; its sheer existence enables it to infringe on copyright law, because if someone prompt-engineers a response sufficiently, the AI will eventually generate something legally questionable, and you cycle back to the previous paragraph, where there’s nobody objectively at fault. Artificial intelligence is, therefore, best left to its own devices, not only so that we may not hinder human advancement, but simply because the mere proposal of its regulation in relation to IP law leads to endless arguments which IP is simply not made to solve, and by extension, to arbitrary decisions which will help nobody and harm everyone.

A Realistic Solution to an Absurd Problem

When one looks into the contradictions and counterproductive effects of IP law, it is impossible not to wonder how it is that this is allowed to happen. It is perhaps a biased view, maybe there’s more to be won from IP law than from getting rid of it (according to a certain tomboyish author). However, the truth is that humanity existed for most of its history without such a system, and yet, we have progressed immensely; even countries where IP legislation has historically, and in modern times, been weak, have been able to see huge technological progress and important innovations, as is the case with China, and to a lesser extent, India and Brazil. That is not to say that a lack of IP law necessarily drove their progress, but it is a clear hint that IP is not the end-all be-all of human creativity and development.

 

After such a long critique, the only logical thing left for me to do would be to propose a solution, after all, what kind of person would I be if I shoot to wound, but not to kill? If I am to be fair, my philosophy already includes a very simple trick that would solve this entire problem almost instantaneously: just get rid of IP law and let the market sort itself out. While this is what most of my peers would likely propose, without compromise, I must be honest with myself and realistic for those in my audience who have not yet become radicalized enough to think that evil cannot be reformed. I absolutely love the proposal of letting spontaneous action take the lead here, but concrete reality is not as friendly with the idealism engraved in my heart, mind and entrails, and it wouldn’t be so friendly with such an idea which would easily get anyone proposing it labeled as either a “communist” or a “corporate-loving fascist”. To put it differently, for IP to be completely erased, we would need a complete worldwide shift in paradigm, and that’s one shift that will never happen if those of us who oppose it decide to stay put while we wait for Richard Stallman to be crucified and revealed to be the second coming of Christ. There are simply too many factors that come into play when we think of IP, all of which work as constraints, both internal and external, which would make just getting rid of any and all IP legislation a completely impossible thing.

 

One of such constraints is the treaties mentioned during the (hardly-abridged-at-all) explanation of IP’s history. If any country today were to abolish even just some core IP-related legislation, as the vast majority are bound by the TRIPS agreement and members of the WTO, said nations would be sanctioned into oblivion, suffer massive capital flight (with everything this entails), and become diplomatically isolated. Not even the most radical parties with any form of power, anywhere in the world, have even remotely cared to propose abolishing any kind of IP, and any of those who have (such as the Swedish Pirate Party), naturally never found themselves winning a single seat anywhere within their political structures. The most we’ve seen have been freer approaches to IP, such is the case with India’s patent law, which refuses patents that lead to evergreening⁶⁶, or former Ecuador president Rafael Correa’s small IP reform, which allowed the production of generic variations of patented drugs⁶⁷, but naturally, these fall terribly short of working as any proper reform of the wretched intellectual property system that rules over most of the world.

 

Given that the idea of simply abolishing IP is very far from reach yet, I decide in its place to propose an updated, heterodox framework, with which to create a compromise between intellectual ownership and freedom of accessing and sharing information. This is in no way a proposal for a permanent solution, but rather, what should be the first step towards a world where just the idea of vanquishing IP stops being taboo. This framework has not actually been influenced by any other authors or thinkers, at least not consciously, but I imagine much of it is far from new, so just consider it a form of parallel thought; the same kind that would probably have me justify myself before a judge for just thinking the same thing someone else did a little too late.

Ending the Patent Monarchy

Currently, most patent laws across the world provide the creators (or whoever was granted the patent) a legal monopoly over the patented invention. These patents have varying lengths, but both in the United States and my native Argentina, this number is 20 years, which aligns with the TRIPS agreement, to which, as said before, most nations are a party through the WTO. Generally, no renewals past this period are allowed, so after 20 years, that patent becomes free to use. However, minor tweaks can practically expand these patent monopolies by 10 additional years (although it depends on national laws), as explained with examples such as that of Insulin previously.

 

My first point to be made is that 20 years of monopoly is a damn long time. Just think that, 20 years before this was written, I was not yet in kindergarten, and nobody had a computer with 100 times the processing power of a Pentium 4 in their pocket at all times. Such a duration for patents can severely delay any potential innovations, let alone competitors selling the same product, right in the midst of the era of most accelerated technological development we’ve seen. This easily leads to stagnation, artificial scarcity, and it makes patent trolls have even more of an incentive to claim patents. For my personal proposal, I say that the maximum length of any patent should be 5 to 10 years for completely new inventions, and 2 or 3 years for any tweaks or improvements over existing ones. This will shorten the delay of any new competition entering the market, and it will work as an incentive for quicker innovation, reducing stagnation and reducing profitability for patent trolls. Additionally, I propose that patents over digital-only inventions last a maximum of 2 or 3 years, which means that any creation that can only be applied to software, for an example, a new gesture-based authentication system, becomes free to copy after that time, to allow open source or free software to implement them before they become deprecated or obsolete.

 

I also propose that patents only apply over tangible inventions. In other words, it should not be possible to patent methods or procedures, thus any new technique for extracting cotton, or any specific use of a natural product, should not be possible to protect through IP law. This will aid ending biopiracy such as the cases involving India, mentioned earlier in the text. It also binds into the problem with intellectual property applying over actions, which is rather absurd (then again, read Kinsella, really). However, this should not apply over machines which provide or enable these new methods or procedures, such as the case of a cotton-extraction machine, or a machine which can process a plant for medicinal use, since these are clear examples of tangible inventions.

 

The third, and perhaps most important, point regarding patents is moving from the current restrictive exclusivity system to a royalty-based model. This means that patents, instead of granting exclusive rights to its creators, grant them exclusive rights over a percentage of the profit generated by the sale of said inventions. Since setting an arbitrary number of this would perhaps not be very sensible, patent-holders should be required to fill in the percentage of income that they require from others for the use of their patent. A maximum percentage could be set, for instance, 50%, or alternatively allow de facto exclusivity by allowing creators to demand 100% of the royalties from the use of their ideas, with said percentage decaying over time until it reaches 0%, at the end of the patent’s life. Under this system, the patent holders would be able to freely profit from their creation, while also being able to profit off others who use it, without fully monopolizing it, but allowing them to keep the majority of the market share by letting them sell their product at a lower market price, given competitors will have to pay royalties to the patent holder. The patent holder would still be able to sue in case the patent is profited from without royalties being paid out, which would simply replace the lawsuits for the invention being used at all. Given that the royalty percentage would decay, eventually the patentor would inevitably have to compete in equal footing with the rest of the people offering their invention.

Protecting Creativity Without Punishing Knowledge

As mentioned before, copyright has a large list of issues, primarily due to the fact that it is in its nature to limit the free flow and access to information. While I would personally prefer for copyright to be entirely destroyed, annihilated, trounced, vanquished, and replaced by copyleft alternatives, this is still as distant for modern legal frameworks to achieve as reaching Alpha Centauri is for space exploration. Therefore, this following section will focus on my proposal to rework modern copyright law, and a Q&A regarding my proposals.

 

Copyrighted media, such as books, films, videogames or music, should be allowed to be shared freely in private after purchase. If I buy a book, it should be possible for me to share this book privately with other individuals, who would not have a right themselves to share it, effectively limiting sharing to the owners of physical or digital copies of said media. Nevertheless, this new framework would disallow publicly sharing such media on the internet or in public (for as long as the copyright lasts). Profiting from the original work by copying it should remain (mostly) illegal, unless specifically allowed by its author, who could collect royalties or sue over perceived copyright infringement. Selling the physical media or any unique digital license which provides access to the work would, however, remain legal and not bound to any payment of royalties. Any derivatives from a copyrighted work would be allowed, such as translations or fan fictions, but these too would be subject to compensation in case of being sold, although their public, free sharing would be guaranteed. The use of copyrighted works for training AI models and other similar purposes should remain largely allowed, and legal grounds for copyright infringement should only arise in the case that any Large Language Model trained on that data shares large portions of those works, verbatim, on demand (which is highly unlikely to happen, anyway).

How would this be enforced, though?

The sharing of multimedia in private is hardly controllable to begin with, since it’d require state agents to invade the privacy of individuals at almost all times (which I’m sure they’d be happy to do if there weren’t 8 billion of us to police at all times). The idea behind this reform is to not criminalize the cases in which people do share things in private, keeping only public unauthorized sharing of a copyrighted work as a type of infringement on the IP rights of the original creator.

How would the scope of private sharing be limited?

This is a fair concern. Theoretically, one might be able to share a book with hundreds of people if they ask you to share it, and you agree to it, at which point, you could hardly call what you’re doing “private sharing”. Now, the truth is that these situations are rather extreme and unrealistic, because of physical and human limitations; you can only copy a tangible work as many times as you can afford it, and answer to as many online requests for sharing as you’ve got time to. Yet, in case of this happening, it is not logical to set an arbitrary number after which “private” becomes “public”, so in case a copyrighted work is copied and shared privately in large-enough quantities as to be brought into legal question, it should be left up to a court to decide based on intent, scope and method of sharing, and whether the person sharing the work openly promoted it, to determine whether it was fair private sharing or public, given this is simply a gray area where arbitrary solutions would be too ridiculous. In such cases, sharing a movie through a chatbot which provides you with it when asked would be a form of “public sharing”, even if there needs to be interaction with a form of mediator first, since in this case it is automated and not different from clicking on a download link.

What about digital ownership?

Many forms of multimedia provide a form of digital “ownership” through licenses (such as software) or by giving unlimited access to an item by linking it to one’s account on a website or application (such as with games on Steam, or music on Bandcamp). Sharing games bought on Steam, software bought from its developer, or music bought from Bandcamp privately (as in, the digital files comprising these forms of media), should be considered the same as doing it with the tangible counterparts of these works; legal, implying you have bought the digital work yourself. Sharing it publicly (piracy), though, or selling a copy of these digital works without authorization, would still be infringement of copyright law, since you’d be profiting off someone else’s work and, on top of that, it would be selling a non-scarce item over which you cannot lose ownership unless its use is limited through licenses, in which case, it would not be different to selling a book one purchased, since you’re selling your access to it.

What if someone makes an indirect profit from sharing copyrighted works?

Say that someone puts up a website that shares pirated copyrighted books, but makes money through advertising. The first issue would be that they’re publicly sharing the works, which would still be illegal, as mentioned before. Nevertheless, if they managed to find a way of indirectly profiting from sharing these works privately, as long as they aren’t making the person receiving the work pay them for it (since that’d just be buying the copyrighted work, directly or indirectly), it should not be a problem, since it’d lead to many cases in which you’d have to arbitrarily decide whether that indirect profiting is legal or not, thus it couldn’t be objectively and consistently enforced. For instance, if I’ve got a website which I advertise for downloading those books, but I also use it to sell my own books, then you’d run into a gray area whereby you’d have to determine whether the profit I made from my own books is thanks to the fact people visited my website to download copyrighted works. In this case, only direct causation should be applied: profits made indirectly cannot be punished due to their nature being entirely up for interpretation and bound to multiple externalities, but profits made through direct transactions in exchange for copyrighted material would remain legally punishable.

How would creators keep revenue if their works can be shared widely?

This, perhaps, is the most valid concern of all. Since any creator depends on the revenue of their work, allowing people to share it to greater extents legally could, theoretically, reduce their income significantly. A proposal would be to trigger all these aforementioned points a certain amount of time after a work is published (for example, a year), so, you’d buy a book, and a year after the book was published, you’d have the right to share it publicly, albeit without the possibility to profit from it. This would allow for profits to be kept when they’re generally at the highest, which is upon release.

 

Additionally, it is important to note that in the modern world, a majority of people no longer actually directly pay for most media. Music is often listened to through streaming platforms, which pay royalties to the musicians or labels. Books are normally read through various subscription services and still purchased due to the fact that people simply like feeling the paper in their fingers when reading. Videogames are almost entirely bought through digital stores, which apply DRM to prevent piracy. Movies are watched at a cinema for the experience, or through streaming. In reality, most media these days tends to have copyright basically sidelined, because the methods used to share them and consume them already provide profits without the need for copyright enforcement (even if it does protect their business models), mostly out of convenience; people prefer paying a fee to watch movies on Netflix or play a game on Steam rather than to pirate it. On top of this, the increasing sophistication of DRM systems and other anti-piracy technology is capable of rendering copyright largely pointless in many cases, as piracy becomes harder through anti-piracy innovation which, ironically, exists to fix the shortcomings of copyright law.

What would creators do once their copyright runs out?

Often, when one proposes the idea of outright abolishing copyright, the first question people ask tends to be: “are you stupid?”. Naturally, after one takes due offense, the next thing they ask is: “what would writers, musicians, directors and others do without copyright protecting their profits?”. This is a great question, because it certainly does touch on what’s practically one of the reasons copyright exists in the first place (at least on paper): to allow people to make a living out of creative works. One could argue that if you suddenly decide to get rid of copyright, then an author like Haruki Murakami, or a musician such as Taylor Swift, would simply find themselves without a source of income, nor any reason to continue writing.

My first answer is rather simple: I think some people have a rather alarming lack of faith in humanity and good will, and that’s likely just projection. My second answer, which is the one that gets to the issue, is that, at least in this case, I am not (yet) proposing to simply get rid of copyright altogether. Under my proposed framework, authors would retain a large degree of control over their own creative works, which they could obviously use to profit. The issue, though, is that just as much as copyright gives authors a reason to write, filmmakers a reason to film, and musicians a reason to music (what do you mean “music” is not a verb?), it also ironically gives them just as much reason not to. For instance, musicians like Vanilla Ice or Gotye made themselves a fortune from just one song, while others, like game developer Markus “Notch” Persson, made himself a billionaire by creating a simple game about placing blocks, called Minecraft. All three of them largely stopped producing more music or games after that (at least for a long time). What I mean by this is that, if any one work is successful enough, it can set its author up for life, allowing them to live off its royalties for decades, without there being any financial incentive for them to create anything new, and perhaps even the possibility that anything new could tarnish their reputation if it’s not as good as their previous work.

 

As I had mentioned in the answer to the previous question, we could have copyrighted works become publicly shareable after a certain amount of time, such as a year, after which the author could possibly see a sharp drop in income. While the concern is understandable, the truth is that, right now, what prevents people from pirating anything they want is not really the existence of copyright law, but the existence of DRM, the limited access to newer works, and just a personal sense of morality. The truth is that, if someone has the money, and likes a creators’ work, it is likely for them to support said creator by willingly paying for their work, even if they can pirate it easily. If this was not the case, platforms like Patreon, Bandcamp or Steam would have long since gone defunct. People just tend to have a predilection towards supporting those they like, as long as they have the financial means to do so.

 

I also think it is rather reductionist to believe that artists and authors live only from the royalties they make out of their works. Most musicians make much more from touring and playing shows than they make from royalties. Many authors tend to make large sums of money by providing workshops, talks, assisting conventions/expos, and carrying out other similar activities. This is similarly the case for filmmakers, visual artists and other recreational entrepreneurs; often, their own image and person becomes greater than their works.

 

On top of that, the lack of copyright does in no way mean that any artist’s work would not be sold. As I mentioned, not only do people voluntarily pay for the work of those creators they like, but many are more than willing to pay a hefty amount for any collectible, special edition of any work they like. Signed books, boxed sets, vinyls, pre-order specials, merchandise and other things often get sold out at light speed, for large sums of money too. Why would anyone pay double for a signed version of a book, $100 extra for a boxed set of a game they can buy digitally and play either way, or $50 for a 7” vinyl with two songs they can listen to on Spotify? It’s simple: because we humans have an undeniable love for anything that feels unique or limited, much more so when it is tangible, which is largely the reason why vinyls and cassettes have not yet died and are often sold out by your average do-it-yourself suburban punk band, or why even web novel authors sell out any print versions of their works, even when they’re not copyrighted.

Share This, Freely (As in Allowed), Freely (As in Free Beer), Openly

I do not care to claim any right over what is written in this document. I recognize these ideas as not mine, but belonging to a collective of people past, present, and future, to the millions throughout history who have share similar stances, and to the many millions whom I hope will one day carry this same torch I now carry (and set myself on fire with).

 

So go ahead and share this, modify this, put it through an AI so that it rewrites it as if it was a medieval theological essay or as if written by Søren Kierkegaard during an LSD trip, do whatever you want with it, but out of nobility, do consider mentioning me as the original author, since I could really use the praise.

 

Thank you for reading (or scrolling all the way down here after skipping the entire text because you didn’t agree with a particular idea).

Footnotes and References

  1. Kinsella, Stephan Norman. Contra la propiedad intelectual (digital re-edition, Spanish translation). 2020, p. 6.
  2. Palmer, Thomas Gordon. “Are Patents and Copyrights Morally Justified? The Philosophy of Property Rights and Ideal Objects”. Harvard Journal of Law & Public Policy 13, no. 3 (Summer 1990). 1990.
  3. Athenaeus of Naucratis. Deipnosophistae. c. 200 AD. Referenced edition: Lipsia, Weidmann, 1827.
  4. De la Durantaye, Katharina. “The Origins of the Protection of Literary Authorship in Ancient Rome”. Boston University International Law Journal 25, no. 1, Spring 2007.
  5. Sendal, René. “Venetian Patent law — 1474”. History Walks in Venice, 2024.
  6. Nard, Craig Allen, and Andrew P. Morriss. “Constitutionalizing Patents: From Venice to Philadelphia”. Review of Law & Economic, vol. 2, no. 2, 2006, pp. 12-13.
  7. Ibid., p. 50.
  8. Statute of Anne, London (1710). UK Parliamentary Archives.
  9. The Constitution of the United States of America. 1789. National Archives.
  10. “An Act to promote the progress of useful Arts” (Patent Act of 1790). United States Statutes at Large, Vol. 1, 1st Congress, 2nd Session, Chapter 7. 1790.
  11. Convention de Paris pour la Protection de la Propiété Industrielle. Concluded March 20, 1883, as revised and amended. World Intellectual Property Organization.
  12. Berne Convention for the Protection of Literary and Artistic Works. Concluded September 9, 1886, as revised and amended (latest Act of Paris, 1971, amended 1979). World Intellectual Property Organization.
  13. Agreement on Trade-Related Aspects of Intellectual Property Rights (TRIPS Agreement). Concluded April 15, 1994. World Trade Organization.
  14. Digital Millennium Copyright Act, Pub.L. 105–304, 112 Stat. 2860. Enacted October 28, 1998. United States Congress.
  15. GNU Project. GNU Operating System.
  16. Creative Commons.
  17. Oxford English Dictionary via Google.
  18. ”EpiPen outrage: It costs about $8 to make a 2-pack, engineers say”. The Mercury News, October 1, 2016 (updated March 17, 2017).
  19. Luthra, Shefali and Anna Gorman. ”Rising Cost of PrEP to Prevent HIV Infection Pushes It Out of Reach for Many”. National Public Radio, June 30, 2018.
  20. ”Generic Truvada costs less than a dollar. Here’s why you’re likely paying much more”. 46brooklyn Research, May 25, 2021.
  21. Gardner, Jonathan. ”Two decades and $200 billion: AbbVie’s Humira monopoly nears its end”. Biopharma Dive, March 17, 2022 (updated January 27, 2023).
  22. ”Patents Assigned to Cochlear Limited”. Justia Patents. Accessed May 30, 2025.
  23. Bowman v. Monsanto Co., 569 U.S. 278 (2013).
  24. ”Over 10,000 people die every day the UK blocks Covid vaccine patent waiver”. Global Justice Now, October 2, 2021.
  25. ”Access to HIV Treatment and Care”. UNAIDS, 2007.
  26. Bollyky, Thomas J. ”Access to Drugs for Treatment of Noncommunicable Diseases”. PLoS Medicine 10, no. 7 (2013): e1001485. July 23, 2013.
  27. William A. Haseltine. ”Global Inequities in Access to Drugs Costs Millions of Lives Each Year”. Forbes, April 28, 2024.
  28. Naturally, I say “in part” here because it must be understood that patent laws cannot be the sole culprit behind such complex issues: bad economic policy leading to unhealthy environments for both local and foreign investment in healthcare is another huge problem, particularly in underdeveloped nations. For a better understanding of this topic, I recommend reading The Mystery of Capital: Why Capitalism Triumphs in the West and Fails Everywhere Else by Hernando de Soto.
  29. To be clear, piracy exists everywhere, even in high-income nations, but within professional settings, it is common for this to be avoided due to stronger implementation of IP law. The reasons for piracy are varied, but one of the main drivers of it is, undeniably, the cost of software, which is in many cases driven up by a lack of competition through patenting. For some statistics on piracy rates by country (from the few which have been studied), refer to A Closer Look at the Countries with Highest Piracy Rates by Bytescare.
  30. For further reading on these topics, I recommend Copyright: a roadblock to education in developing countries? by Noha El Labban, The impact of digital copyright law and policy on access to knowledge and learning by Tana Pistorius & Oridachukwu S. Mwim, Copyright and Inequality by Lea Shaver, Unlocking the Potential of Copyright Limitations and Exceptions (L&Es) by Faith O. Majekolagbe, and International Copyright Law and Access to Education in Developing Countries : Exploring Multilateral Legal and Quasi-Legal Solutions by Susan Isiko.
  31. Donnelly, Claire, and Meghna Chakrabarti. ”The Internet Archives is in danger”. WBUR, January 7, 2025.
  32. Vaziri, Aidin. ”Musicians rally to defend San Francisco’s Internet Archive against $621 million lawsuit”. San Francisco Chronicle, December 10, 2024.
  33. Marasigan, Marc. ”Ubisoft Faces Backlash for Making The Crew Completely Unplayable After Its Closure”. MMOs.com, April 25, 2024.
  34. Pitts, Lan. ”Ubisoft Is Facing Lawsuit After The Crew Was Removed from Platforms”. GameSpot, November 12, 2024.
  35. ”Defunct Online Videogames”. TVTropes. Accessed May 30, 2025.
  36. ”Disney’s Snow White ends in disaster: A $215 million flop that shakes studio’s future”. Marca English, April 27, 2025.
  37. Bergin, Lauren. ”DragonForce’s Herman Li banned from Twitch amid DMCA drama”. Dexerto, October 27, 2020 (updated November 2, 2020).
  38. Here I say “far-fetched” because I believe we are yet to see any instances of an artist being fully erased or censored from the mainstream through copyright just for “wrongthink”, however, artists such as Ted Nugent or Morrissey have had their music deemphasized by their labels and copyright holders for controversial political opinions which have been labeled “far-right” by the media.
  39. NTP, Inc. v. Research in Motion, Ltd., 418 F.3d 1282 (Fed. Cir. 2005).
  40. Kelley, Rob. ”BlackBerry maker, NTP ink $612 million settlement”. CNN Money, March 3, 2006.
  41. Eolas Technologies Inc. v. Microsoft Corp., 399 F.3d 1325 (Fed. Cir. 2005).
  42. Rose, Mike. ”Lodsys Files Disputes Against App Store Developers”. Gamasutra, June 1, 2011. See also Lodsys, LLC v. Combay, Inc. et al., No. 2:11-cv-00272 (E.D. Tex. Filed May 31, 2011).
  43. Apple Inc. v. Samsung Electronics Co., Ltd., No. 5:2011cv01846 (N.D. Cal. Filed Apr. 15, 2011)
  44. Albasoos, Hani, and Nabil Al Musallami. The conflict between Apple and Samsung over patents and copyrights. Bussecon Review of Social Sciences 3, no. 3 (2020): 3-5..
  45. The Apple Patent Fight Between Apple and Samsung: Interviews with Korean and Korean-American Attorneys. University of California, Irvine School of Law.
  46. ”When Patents Attack”. This American Life, National Public Radio, July 22, 2011.
  47. European Patent No. EP0617119A3. Co-extraction of Azadirachtin and Neem Oil. Filed March 17, 1994.
  48. Anveksha. ”A Case of Bio Piracy: The Neem Patent Controversy”. Office of Advocate Partap Singh, July 12, 2024.
  49. U.S. Patent No. 5,663,484. Basmati Rice Lines and Grains. Filed July 8, 1994.
  50. Rai, Saritha. ”India-U.S. Fight on Basmati Rice Is Mostly Settled”. The New York Times, August 25, 2001, Section C, p. 1.
  51. U.S. Patent No. 5,401,504. Use of Turmeric in Wound Healing. Filed December 28, 1993.
  52. Bhowmick, Anusree, Smaranika Deb Roy, and Mitu De. ”A Brief Review on the Turmeric Patent Case With Its Implications on the Documentation of Traditional Knowledge”. NDC E-BIOS 1 (2021): 86.
  53. U.S. Plant Patent No. PP5,751. Banisteriopsis caapi (cv) `Da Vine`. Filed November 7, 1984.
  54. Bosse, Jocelyn. ”Re-examining the “wild” story of the ayahuasca plant patent”. Intellectual Properly Quarterly [2024] Issue 2: 136-162 (p. 25 in cited PDF).
  55. Schudel, Matt. ”Accomplished, Frustrated Inventor Dies”. The Washington Post. February 26, 2005, p. B01. See also ”Biography:Robert Kearns”. HandWiki. Accessed May 30, 2025.
  56. ”Wright brothers patent war”. Wikipedia, the free encyclopedia. Last modified March 31, 2025. Accessed May 30, 2025.
  57. Good Morning to You Productions Corp. v. Warner/Chappell Music, Inc., No. 2:2013cv04460 (C.D. Cal. Filed June 13, 2013).
  58. RCA Patent Law Department. Files on Armstrong v. Radio Corporation of America and National Broadcasting Company, Inc., 1948-1959. Accession 2464-80, Identifier 2464-80-VI. Records of other RCA divisions (Collection 2464-80). Manuscripts and Archives Repository, Hagley Museum and Library, Wilmington, DE.
  59. Browne, Malcom W. ”Papers Tell Tragic Story of Man Who Invented FM”. The New York Times, May 21, 1978, Page 13.
  60. The Prisoner’s Dilemma is a concept in game theory which seeks to demonstrate how rational individuals, acting in their own self-interest, can lead to a worse outcome for everyone involved than if they had cooperated. In this context, to provide a more concrete example, imagine two companies: Ruritania Tech and Waldavia Solutions. Ruritania Tech creates a new digital storage device, and considers not to patent it, so that Waldavia Solutions (and everyone else) can benefit from it, by producing it, competing over its market share (thus driving prices down), and improving it, but Ruritania Tech also considers the possibility that, if they don’t patent it, Waldavia Solutions might decide to do it instead, thus stealing Ruritania Tech’s invention. Fearing exploitation of IP law by Waldavia Solutions, Ruritania Tech finally decides to protect their creation through IP law, to avoid having it taken away from them by their competitors.
  61. In this case, I’m being somewhat reductionist for clarity’s sake. While the broad concept of the steam engine wasn’t patented in a way that prevented its development, there were nevertheless some patents (such as James Watt’s) which did restrict it. The point I’m trying to get to, though, is that the initial spark for these revolutionary ideas wasn’t necessarily IP-driven in the modern sense, nor protected using legislation comparable to what exists now.
  62. Authors Guild, Inc. et al. v. OpenAI, Inc. et al., No. 1:2023cv08292 (S.D.N.Y. Filed Sept. 19, 2023).
  63. Silverman et al. v. OpenAI, Inc. et al., No. 3:2023cv03416 (N.D. Cal. Filed July 7, 2023).
  64. Kadrey et al. v. Meta Platforms, Inc., No. 4:2023cv03417 (N.D. Cal. Filed July 7, 2023).
  65. Andersen et al. v. Stability AI Ltd. et al., No. 3:2023cv00201 (N.D. Cal. Filed Jan. 13, 2023).
  66. India. The Patents Act, 1970 No. 39 of 1970, as amended, sec. 3(d).
  67. Ecuador. Decreto Ejecutivo No. 118. Published in Registro Oficial [Quito], Año 1, No. 67, November 16, 2009, p. 3.

YOUTUBE transcript:

because uh Okay go ahead but uh so this is Stephan Gonalo i’m going to do this as uh podcast

K467 and I’m talking to Wani from which country are you from argentina from

Argentina uh about intellectual property you had written uh a draft article on

IP and it’s about 25 pages and I I I commented on it briefly but I got Bro

Grock to look at it gave you some other articles to look at and then

we can talk about it so let so go ahead tell me what your your view is and what you want to discuss so um I’ve got a

rather standard libertarian view on intellectual property and it was shaped

in great part by your works but um much of it actually comes from a a lot of

personal thinking and a more of a consequentialist view on the issues

caused by intellectual property so my criticism of intellectual property is

not necessarily so much on the philosophical side of it but on the I I guess you

could say it is a bit more utilitarian in nature well hold on before we go there let me ask you a question so um

are you a student or a scholar or what’s your um I’m actually just a programmer

i’ve already I I already got two um technical degrees from university but um

I’ve never done anything in this sense so I’m more of a like

a you could say a an afficionado in this i’m not necessarily a star or anything

like that or would you say you’re a libertarian or anarchist or um um I consider myself a

voluntarily just do not necessarily say anarchco capitalist mostly because I

perhaps have a few teological differences with the average anarchco

capitalist and because basically that word when you say you say someone yeah I’m anarcho capitalist they look at you

like you just said you’re a Nazi no I just want to see where you’re coming from are you Austrian sympathetic misian

sympathetic yeah um can say um that much uh knowledgeable on economics um that’s

not necessarily my strong point yet but yeah I’m Austrian um Roberian

um you could say that I’m very very much standard in that in that sense um

probably have a bit more of a left libertarian leanings in some stuff

that’s fine okay hold on just a second let me um I’m going to pause this i’m going to I need to grab a new battery

for my fan give me just a second okay back um and are you are you a fan

of Malay or what’s your situation on your views yeah yeah pretty much so uh

or at least on most of his stuff like you know I absolutely love his reforms

in terms of intellectual property he hasn’t done so much yet but he has for

instance um we used to have all our society for

artists and musicians and composers the one that writes everything um you know

the one that manages everything related to copyright over music and phonographic rights um he had removed a few

regulations that for instance made you if you let’s say you own a business and

you play music on Spotify you were supposed to pay royalties to the artists

in order to play that music in a public space and that even you know that could

even apply if you played your music at your home um loud enough for people

outside to hear mhm so naturally he removed that it was one of the few IP

related um reforms he has done so far from what I know

and well unsurprisingly a lot of people especially most um mainstream

musicians had a well they had a complete meltdown over it so far nothing so bad

has happened so in that sense yeah I absolutely I wasn’t Yeah I wasn’t sure

what his views on that were uh I think someone told me he had read one of my books but who knows um I’m pretty sure

he must have read some of your books since he is like from what I know of him

he is absolutely reading all day so I’m pretty sure he he knows you well um but

the thing is that his reforms are much more pragmatic uh he hasn’t touched much on IP because

I guess he has a more random view of how to deal with government right

well maybe if you can find an article that explains what he did um on the

copyright issue and what controversy it had I could uh I could take a look at that but anyway let’s go ahead tell me

what any anything you want to discuss or what questions you have all right so I had written this essay and it’s

basically pretty much an argumentative essay that um I wrote when I began

thinking a lot over my own criticisms of intellectual property and it is

something that I hadn’t necessarily seen a lot of people discuss directly because a lot of works on intellectual property

um are more on the philosophical intricacies and the problems with it not

so much as the re as what I perceive to be the real world problems not to say that you haven’t written about it you

have written extensively about you know paid trolls and the problems with big farm and

pants and that stuff but um I was thinking a lot of this and I wanted to

organize my thoughts so I wrote this essay which is title ideas for free a case against intellectual property and

one of the things that I did which is something that I would like to discuss

with you about principally is I began thinking a lot about intellectual

property as it relates to artificial

intelligence you see artificial intelligence is still very much um new

in the sense of how it has entered the mainstream it has existed for a long

time sure but right now we got generative AI and you can simply go

ahead um open up chat GBT or Grock or any other

model you can input a prompt and you suddenly are going to have probably you

can get an entire essay on any topic or an entire new story or you can even have

a whole poem a whole um lyrics book written by an AI a whole song a movie

script whatever you want you can have it generated by an by artificial intelligence right now and I realized

hey how how does IP specifically copyright but but also patent rights

come into play here because recently this is something that I mentioned in my

article or ASM more like um there have

been a series of lawsuits regarding IP

um over artificial intelligence models you got auto skills versus open AI

Silverman versus Open AI and meta and these people mostly others um file law

file lawsuits uh to these companies because they said hey your model is capable of generating

stuff that’s very similar to what we wrote and uh on copight

over and so this brought forward the question of like how do you

determine what um I mean how do you determine that the AI generated

something that’s similar to what you own copyright over to your

works if you cannot necessarily prove that they were trained on your works

illegally illegally as in without a proper license and isn’t this just you know a

case of periling um carried out by a

computer and who and who do you blame who do you blame the user who made the problem do you blame the people who run

the AI model and train the AI on that data and even then if they didn’t train

the AI on your works if you engineer a prompt long enough with enough um

specifications and and stuff you’re going to eventually have the AI generate something that is legally questionable

and so that’s well let let me let let me interrupt with a couple things here that you might not be familiar with there’s

two issues you’re kind of blurring together so one is the issue of authorship um so let’s just suppose

let’s forget about the copying but let’s just suppose you own a computer and you program it to um to just generate things

and it generates a pattern like a a picture then the question is is there a

copyright in that and who owns it uh obviously the machine can’t own it because machines can’t own things um but

it might not be owned by anyone there may be no copyright on it there was there have been some disputes about

these example where someone some photographer or some person gave his camera to a monkey called him a [ __ ] and

he took some pictures which became internet famous on the internet and the courts ended up holding that there’s no

copyright because there’s no author now on the other hand if you you know If you

throw your camera up in the air and it takes a picture and you catch it you can say you you you took that picture if you

set up a stationary camera like on your building and it it just takes a movies of the parking lot for security is there

a copyright in that is there anything artistic or original about that i I don’t even know the answer to that but

there are questions about who the author is and who whether there’s a copyright

in something so that’s a separate question so the question is is if AI generates an artwork is there a

copyright in it or not and if it’s totally original I think the argument is either that the person who programmed it

is the author like if you give it enough prompts to tell it what to do you could argue that you have a copyright in it

similar to the way photographers have copyright in photographs even though they’re in a way factual representations

of the world and because the law the law decided long ago that there’s a minimal enough

uh original contribution on the part of the photographer in telling the camera what to do the you know the um the the

shutter size the the the lighting that’s enough to give it

originality but if it’s if there’s no contribution at all like you just set the camera up and it records for for

days a parking lot uh I don’t know if there’s copyright in that and I don’t know if how much the author now of

course there’s no objective answer to these things and this shows that there’s something wrong with copyright law yeah this question is so the other question

is though the other question is this you have to understand that copyright covers a group of rights one of which is

copying which includes what’s called literal copying or exact copying but it also includes sufficiently similar

copying so like if you take someone’s book and you you you change the wording a little bit it’s still a copy because

it’s close enough it’s substantially similar or if it’s a literal copy it’s a literal copy but there’s another type of

copyright which is called the derivative work and that is um when you make

something that’s not a copy but it’s it’s based upon the original idea so if you make a translation of a book or if

you make a movie based upon a novel those are not copies but they’re still a violation of the copyright because the

right to make a derivative work is also one of the rights so the question would be if you program an AI to generate a

new work and it uses um other people’s copyrighted material um is that a

derivative work that infringes on the rights of the holder um and then the other question you mentioned this thing

about inevitably the AI will generate something that will get into trouble i don’t think that’s necessarily true uh

if you just have a an AI randomly generate things um um and it’s not based upon training

data if it happens to be the same text as Atlas Shrugged or a novel it would not be a copyright infringement

because it wouldn’t be a derivative work or a copy it would just be um the same

idea the same pattern by coincidence um in fact there’s one one uh copyright

researcher Damian Real who he did an experiment where he programmed um he

used a computer to program all the melodies of music and he he he thinks that that would be used as an offense in

copyright infringement he’s kind of confused about the law but it was that kind of little trick he tried to play um

but I think the more interesting question is this when to in real world applications you

have to train the AI and to train it it has to have a lot of data and that means

it has to see have real world data which is scanning all the books all the websites is scraping all this data the

question is is that is that scraping which requires temporary copying is that

itself copying that violates fair use or copyright law so that’s the question and

there’s uncertainty about that and there’s lots of lawsuits that are coming which are threatening to chill in fact

I’ve tried I’m sure you’ve tried if you ask Grock or chat GPT to um to give you

some quotes from some novel it’s it’s sort of paraphrases or it’s limited in what it does and I think that’s because

the the programmers are are intentionally hobbling AI to try to

avoid copyright infringement uh which is which is a shame it means it’s going to distort and impede how efficient and how

good AI technology can be so we already are seeing copyright have an effect on

on on how good AI can be so there is definitely a tension or a conflict

between AI technology and between um protecting copyright you can’t have both

but I have Yeah yeah i wrote rather extensively about that like this I it’s

this part of the essay is called IP in the AI era or IP in the AI era or the

revelation of a contradiction that’s what I titled that part and I wrote

extensively about it i explained to whoever is reading that essay that hey

um AI needs data to be trained these data sets are often not picked

like not handpicked they usually are like many terabytes of different stuff

you know um books images videos internet comments tweets that sort of stuff

and naturally you’re bound to have copyrighted stuff in that you know like

a lot of um AI models are trained on for instance tweets or Reddit comments or

forums blogs all this stuff and inevitably you’re going to have like a

lot of Reddit comments which directly quote some copyrighted

book and you cannot really go ahead and check every single one of those millions

and millions of comments which would probably take a person thousands of hours just to make

sure hey this might be copyrighted so we cannot include this specific part of

this data set impos it’s impossible it’s impossible yeah it is impossible but the problem is that for regulators this they

generally don’t care about these implications because they’re going to see that there is a copyright violation and they’re

going to they’re going to go “Hey well it’s your fault it doesn’t matter if it

was not sensible for you to go ahead and

filter.” They’re right it is your fault it’s your fault because the law makes it illegal for you to copy things that’s

copyright law yeah that’s that’s the problem and I do mention it and yeah

well you were saying that if the you were asking if the AI generates a new work is it

um is it a derivative is it a copy well I guess there is no real answer to that

because AI at the end of the day it’s going to generate stuff based on what it

is trained on if you don’t train an AI on anything then it cannot generate anything at best it’s going to be like

monkeys banging on a typewriter and even then I mean imagine if monkeys banged on a typewriter for

millions of years you’re going to eventually have them um copy all the works of Shakespeare or because they are

in the public domain let’s say well they would be copying no they wouldn’t be copying well not copying it would be

completely random yeah independently stumbling across the same pattern that’s true yeah well but AI technically well

not technically but in a practical sense it sort of does that because AI works on

pro on a probabilistic model so it just generates and hold on it works on

training data too so it’s not doing it randomly that’s not how it works so in the real world it is based upon it’s an

oversimplification but um you’re going to eventually probably have an AI

generate something by sheer chance that might be Yeah that’s not the problem

that’s not the problem the problem is that it’s it’s it’s actually copying like if it goes on the internet and

scrapes content or that’s a copying it’s doing a copying and then it uses that data to train its neural net programs

and it’s the copying right there that’s a problem and then it’s what it produces that’s a problem i mean look if I go to

Google Translate right now and I type in I I feed it in um

uh a Harry Potter novel and I say translate this into Spanish it produces

a work which is a derivative work of the original and if I publish that I’m

violating copyright law it’s not it’s nothing to do with how I did it it’s it’s the it’s it’s it’s the back yeah

that’s but that’s also something that I do touch on and I say I I I’m going to

cite the example I I made so um quoting from my essay imagine you ask a large

lang a large language model to generate a particular story about a young guy with glasses who goes to a school for

magicians and then finds himself trapped in a battle royale death match with other students and at some point one of

the characters also happens to explain to the protagonist the intricacies of wealth and income inequality in the

United States and Europe since the 18th century then you share the story publicly the AI well not the AI but um

the content of the AI might be infringing on the copyright of multiple different outers by creating that story

so by sharing that story online you might be trespassing on the copyright of

all these people right yeah infringing infringing copyright that’s correct you’re you share you share it so you

could say well this is your problem because you generated it right but if if

you go to court I mean you’re you are the legal legal expert here um I’m mostly talking from from what I

understand about this um if you go to court then

um you’re probably going to get yourself in legal trouble because you were the one to share that yep but then I think

like okay you shared it sure you may not have any incidents if you didn’t know

that it was um infringing copyright but that’s still your problem it’s not that

that’s no there’s ignorance is no defense um it’s a Yeah right ignorance is a strict liability crime which is

also one of the problems that I have because it basically criminalizes parallel thought

no it doesn’t no that’s that’s a confusion it doesn’t it doesn’t uh because copying

So you’re getting confused with patent law is a you’re right about patent law

no copyright you have to have so so let me explain something in in the way the law works in order to prove copyright

infringement you have to prove that you copied something which requires you to prove that the the copier had access to

the original because if he didn’t have access then maybe he did come up with it on his own it’s just that it’s extremely

unlikely and if you can find an exact duplicate copy of a novel it’s so impossible you had to have access and

the same thing is true with with AI if AI generates a story which is very

similar to another work and you know that the AI was trained upon billions of

pieces of data which included that work then you can prove that it probably had access and that’s the reason the the new

story is too similar so it doesn’t it doesn’t prohibit parallel or or original

um similar ideas it but it what it does is it prohibits you from creating a derivative work based upon the original

idea if you had access to it but go ahead well but then my question to that is I mean first of all I was thinking

more of I understand parallel thinking in terms of patence i did also mention that um I was perhaps thinking of peril

thinking more in the sense of generated a work that is very similar to one that already exists because then that is in

many many ways um sometimes considered plagurism and it can lead to some legal

troubles but beyond that you said that um you need to have access to the work

right so in that example let’s say that I asked Grock to generate that story but

I did not necessarily have access to those books that I’m referencing but but Grock had access yeah but Grock had

access but how do you know that Gro had access because then if um if um JK

Rowling and other authors like [ __ ] or I’m not really sure how his surname is

pronounced he sends me to court over that story then they’re going to have to have X AI

to show the data sets they were training their AI on correct well well and

they’re going to have to find those works there so so let me explain something up until today the way the law

worked is it’s not dealing with massive data sets and all this so what the way the law works for copyright is it has

this test and the test is if the if the new work is very similar to the old work

and if you can show access then then the court can conclude that there was um

there was copying but that test might change under AI because you might not be able to easily

in other words you you might the law might change where they presume

access if if the new work is too similar to to an old work and if the AI had

access to the entire internet or something that the burden of proof might switch from the from the copyright

holder to the uh to the defendant um the defendant to prove that they did not

access data so I could see the law changing to permit copyright to still

work in a world where AI is doing a lot of the the infringing that that could change that’s my guess um yeah but then

you got you know the lawsuits I mentioned and well this is perhaps something you may be better fitted to

answer um so you got you know these authors saying hey OpenAI infringed on my

copyright because they train their data model on my works now the for that to

come up I mean for that to really um matter you’re gonna have to have your

you know the AI generating stuff that is similar to your works going public and

so that you’re so then you’re going to notice that because then you I asked this question you know like you got to

train the AI sure and training the AI is in practical terms as long as the AI is

being trained and not necessarily sharing and generating your works or

anything um very similar to them it’s not different to a person reading your

books in order to learn or in order to get inspiration you you could argue it’s

qualitatively different argu but um what I’m trying to get to here is that well I

mean you could argue that somebody’s is infringing on your copyright by the way

they’re using your books but they are doing so privately i mean if I will go ahead on my computer and begin training

my own AI model hold on hold on let me let me mention something there’s private

private copying is still copying if if if I have a copy of the Harry Potter novel in my house and I put it on my

Xerox machine i mean my photocopier and I make a copy i’m I’m infringing copyright right there because I’m making

a copy now no one can it’s hard to detect it but it that is technically it’s it’s not just the publication of a

of a copying work it’s the copying which is the which is the crime yeah which is one of the things that I find completely

crazy about there’s about a thousand things that are crazy about I mean yeah there are a lot of those but like the

fact that you can um get into legal trouble over this it’s mindblowing like

imagine you go back in time and you say to somebody like the first liberal

thinkers you know you go and tell them “Hey you know well the other day I copied this one book and I got the FBI

knocking on my Hold on hold on hold on i want to tell my wife to to share um

all right i’m bad bad go ahead right okay so as I was telling you um it’s

mind-blowing and when it comes to AI well I’m I mean I’m a software developer

web developer so I’m very much into this it’s extremely like we need to

understand that AI generative AI specifically is perhaps one of the most

revolutionary if not the most revolutionary invention in human history like this stuff is going to bring us to

the next era of technology it’s like the it’s like the modern day industrial

revolution and as it stands right now we are very lucky because in terms of law

and regulation and all this um AI is pretty much the wild west and unlike

what many people think the wild west actually kind of worked even if it didn’t had no formal law um sure it has

its problems and we can both see that like you cannot right now AI has gotten so realistic in generating videos that

you can literally have people um being like the whole likeness of people being

copied which can be extremely misleading and that’s and that is a problem one of the things that I mentioned in my

essay is like there’s likely somebody right now as we speak sitting in some

office somewhere maybe Washington maybe Berlin maybe London somewhere in the

world going hey you know like this AI problem like we got a problem here not

only in terms of copyright but in terms of ethics they’re going to say how can

we make sure that AI is not being used to spread misinformation or um promote

violence or um scam people and then I go like okay so these people and there are

many people um that are not politician bureaucrats but um civilians so to speak

like the average person has says hey we need to regulate AI right now because

this is a problem and if we look at history um there were many industries that had

like a boom a golden age they got regulated and boom they they died like that or not

necessarily died but they got um they ended up being completely deacelerated

in terms of growth and progress and the problem with AI then is okay so you

regulate this and you’re pretty much given the state the capacity to use AI

however they want because the state never follows its own rules that’s that’s like a given that a lot of people

seem to not want to acknowledge but the state doesn’t follow its own rules um so

you’re gonna regulate AI and then on ethical grounds like okay so you we’re

going to regulate this because it’s unethical to use it the way people use it and the state is going to use it the

way people use it yeah let me interject for a second some of these concerns are not IP related of course it is true that

state regulations distort and impede technology in a variety of ways one of

them is the regulations you’re talking about and copyright is another one now up until recently I’ve always said that

patents are worse than copyright because patents um impede innovation and that is uh

directly uh uh uh harming the the future

of humanity because it slows down the accumulation of of knowledge which is how we become richer um and copyright is

is a second because copyright lasts longer but and copyright distorts culture which is bad but it what in

today’s world the most one of the most important tools that we have to fight the state is communication on the

internet and with encryption and cell phones but copyright is being used to

threaten freedom on the internet so I always said that the biggest harm of copyright is that it can threaten

freedom on the internet but now with the rise of AI I’m starting to think that there’s a second and maybe even bigger

harm to the extent that that copyright distorts the development of AI that

might be the biggest single harm of of copyright even bigger than threatening internet freedom and it might even be

worse than um than patent law although patents also affect the development of

AI technology but go ahead yeah it is it is what I also mentioned that I mean

copyright is going to have a massive and I mean a massive hindrance on the

development of AI because the moment that authors and musicians and filmmakers and every other copyright

holder out there um realizes that AI models are being trained on their stuff

and that AI eventually is generating stuff based on their works even if not

direct correctly they’re going to say “Hey this is undermining my copyright.” Um well my this is undermining my

copyright and they’re going to go ahead and you’re going to have these huge

lawsuits and that’s going to be I I don’t want to say it’s going to be the

death of AI because I’m pretty sure I want to believe too that the law systems

are going to see the training of AI using this data as fair use or at least

I want hold on it’s possible I mean you may not be aware But there are some countries like I think in the UK there

has been a bill proposed maybe there’ve been a bill proposed to to relax the fair use rules but of course all the

artists are fighting that and I don’t see that happening in the US anytime soon and even if you do it it will only

be limited and it will it and what’s going to happen is you’re going to have these big AI companies like Chat GBT and

and and and Grock they’re going to go to the big providers of information like

newspapers and publishing industry and they’re going to make a deal and it’s going to increase the cost but it’s also

going to make the AI worse because now the AI can only use sort of limited sets of data instead

of everything it should be able to use everything the mere fact that it can’t use everything or it has to pay a fee is

going to dissuade small startup companies from doing AI only the big ones will be be able to do it so it will

tend to cartilize or or cause a monopolized set of AI industries the

ones that can afford to pay the license fees and the data will be worse because

it won’t be the full set of data it needs to do training so that’s another problem I see with the movement even if

fair use rules or fair dealing rules in other countries are are relaxed um but

let’s hope so but the point is you cannot have copyright and AI you just

can’t have both yeah there’s a there’s a I think that Yeah I think that IP um is

actually like it runs into a problem not only copyright but appearance too and I’m probably going to get into that in a

bit um it runs into a problem with AI because what is I mean so as to like to

simplify um IP was like an invention in the modern sense that tried

to exalt the creativity of people saying hey well you made the mental effort to

come up with this particular idea so you should have rights to monopolize it yeah

what happens when the AI is the one generating that stuff and how do you prove that some that a current new idea

something like if I write a book or if I come up with an idea how do I prove that it is not made by AI or at least not

assisted by AI and how do I like I said like I said but that’s a different issue and that the law is going to have to

adapt there too just like you did with the monkey copyright case and you’re what you’re going and just like the law had to come up with rules to decide

whether or not photographers get a copyright in photographs i I think obviously what’s going to happen is if

you’re a researcher and you use AI as part of your tools like a computer to

come up with with a patentable invention and you file a patent on it then you will get the you will get a patent on it

and then someone will challenge it by saying no one should have a patent because you didn’t invent it Grock

invented it but then the law will sort that out one way or the other but it will just it will just re it will result

in extra cost and uncertainty and it will result in some IP anyway so

it’s that’s what’s going to happen yeah but then the the problem is how do you objectively create laws that can deal

with this because it’s going to be different in in each case you you cannot

really have an objective way to say hey well I can I’m going to regulate AI

based on these uh rules of play okay you’re going to regulate it and it’s not

going to you can’t be objective because the law is not based upon anything objective it’s it’s arbitrary and vague

and unjust i mean yeah that’s the problem now now the the law can draw arbitrary rules and that’s what they’re

going to do but they won’t be just and they will to the extent that they favor IP rights it will impede and harm

technology there’s just no way around that that’s what’s going to happen that’s the problem with IP there’s no

way around it yeah and I mean you’re gonna we’re gonna

as I said we are very much um lucky yet so far that we haven’t had AI be

overregulated or regulated at all really and something that a lot of people are

completely unaware of apparently is that AI doesn’t even need any sort of regulation there’s no even I don’t think

that there is even any sort of actual argument saying hey we need to regulate

AI it’s not like um the 1930s or 1940s aviation industry where you had people

saying hey we need to regulate aviation because it is it is dangerous no it is

not i mean this is not the problem with AI and AI has a particular set of ethics

that every single person engaging in developing AI has to adhere to because

it is a professional standard i mean the in the industry is pretty much self-regulating even the

issue of um AI videos being used to replicate the likeness of people and use

that for a scam or misinformation ai companies are working hard to make sure that doesn’t happen i

mean AI companies are pretty much trying their best to keep themselves out of the

embrace of bureaucracy which is going to absolutely [ __ ] them down to the ground

through regulation but I’m afraid that this is not enough because you’re eventually gonna have the

state go ahead and regulate this either on the on um based on some edit grounds

or whatever or like to maybe perhaps from some regulators who might have some

good intentions as if but most likely um

the state is going to regulate that because it’s first of all it’s going to be useful to them Because AI can be

really be used by the state to for surveillance for

um economic calculation in of in those economies which are more like command

base like China and because AI can also be used to

undermine the state i mean well what do you need by the way just be to be clear the state is

already regulating AI because it has copyright there are copyright rules already i mean copyright is already

regulating AI technology that’s happening now yeah well what what

perhaps what I mean by regulation is like more direct regulation that directly addresses AI and that’s a

problem i mean if I’m going to cite um Albert Knock here at the end of his book

our enemy state he says “What can we do against the estate expanding?” Well nothing and I would say that he was a

bit of a bit pessimistic on that but if we look at it like how can we actually

go ahead and sit here and think of a way to say hey the state shouldn’t be

regulating AI because it’s definitely going to use it for its own purposes

while they won’t play by their rules like they’re going to impose rules on us but they’re going to use it

however they want and that’s one of the main issues I got and m maybe moving now

out of the AI thing you um mentioning how copyright

um has a cultural impact ai I mean IP is censorship i’m that’s the best way I can

put it ip is censorship and I want to bring up two particular cases that more or

less prove this and that’s um they are mostly related to

Disney and so you know Disney a lot of to what to what Disney yeah um Disney is

very much known for first of all copyrighting the first version of M

Mickey Mouse for like almost 90 something years something like that like

the original version like the very first cartoon of Mickey Mouse only entered the

public domain like a few years ago which is completely crazy because um what they

did with Mickey Mouse is basically copyright eing because the current Mickey Mouse

shirt is very different to the one that came up in the 1920s but it it is a character that was

created when 1928 Mickey Mouse and it is still copyrighted mostly

yeah well the early the early films I think just entered the public domain but that’s just the early black and white

ones yeah well um the black and white ones had like that first depiction of

Mickey Mouse it is in the public domain but the rest is not even if Mickey Mouse is as a

character well almost 100 years old not not only that um trademark law is

another type of IP right so there’s still trademarks on like the mouse ears and so IP is not going away just because

some early Mickey Mouse movies that are 100 years old now have entered the public

domain it’s there’s a long tale for copyright and there’s trademark and there’s uh trademark as well so like

let’s remember something here that Disney tried to um

get intellectual property rights over a musical scale i as a musician cannot fathom this

owning not a piece of music not a composition but an entire musical

scale this would imply that every single song every single composition that is

written on that scale would be um in technical terms

infringing on the intellectual property of Disney thank god Disney never managed

to happen but imagine if that would have happened and what I was thinking of then in terms

of hold on but but it does happen in limited ways so for example you can have a trademark on a on a very short tone

you know like the sound tadum on Netflix is probably trademarked and and um and

you know there are songs like a recent lawsuit from a few years ago there was

this company that claimed the rights to the song Happy Birthday yeah um Chapel Warner Chapel I think it was yeah yeah

and they went around for years extorting people to pay um royalties or or movies

or TV shows would just make up or or restaurants would would come up with a new song so they wouldn’t infringe

copyright and finally the copyright was overturned but it took years and millions of because that song was

actually written in the late 19th century it should have been in the public domain yeah but the point is

copyright is vague and arbitrary and you never know and you can use it to bully people even if you shouldn’t have a

copyright yeah and you’re going to get these things that’s one of the main problems

that about it and as I was telling you that’s not the only problem the one of

the problems that I noticed is that copyright is capable of cultural

er erasure and it can be used by the state to collectivize individuals and

smaller communities into like this greater sort of um community that the

state wants to create where everybody is equal um in terms of how they relate to the

state everybody is is supposed to be I I guess you could call it like this

sort of erasing um regional and personal

and ethnic identity to create a national identity

in in a way that legitimizes the power of the state because well modern nation

states are mostly supposed to be based on the concept that a people that lives

in a territory and shares a common culture and common history is given the

capacity to rule itself and this is undermined when you’ve got you know for instance in the

US you’ve got um all these indigenous American groups which claim their own um

lands as their own territory like it is it does not belong to the US but to them

and you then also got um a divide between for instance the people who live in Texas who might not necessarily be

very fun of the federal government while some others who live for instance in New York might so you end up having this

problem where you can use copyright to erase their um culture in a way that

dilutes them into the mainstream into the general idea the state wants to

people to have of themselves and two um good examples I can provide of this

relating to Disney are Pocahontas and Mulan or Muan i’m not really sure how it

is pronounced in English but uh Pocahontas was a native woman i think

she was of um what today is Mexico and

she was literally forced to convert to Christianity and to marry uh John

Rolf entirely for propag propagandistic purposes to legitimize colonial rule

over indigenous Americans groups and what happened to her in the Disney

history um she was literally made to be like

this hero I guess um it was turned into this romantic story which completely

sanitized what it was done to her and her people by um the colonial rule of

that time and the thing is a lot of people don’t realize Pocahontas actually

existed they think she is a decent story and the same with Mulan who was um

according to sources she actually did exist and

uh well she was like she represented confusion values and was like a more of

a in Chinese culture she has more of a trai the traditional role of a woman and

the version by Disney basically downplayed this these traditional

confusion values and repackaged her and her life as some culturally grounded

legend about some hero which completely powers the western if and if you try to

publish the true story of Pocahontas you might get sued by Disney and so you

might you might win but it would be expensive to fight it so it would dissuade people from telling the true

story because they might want to show they want might want to show clips of the Disney movie in their documentary or

whatever to show the difference so but all you’re doing I think is pointing out all the obvious negative effects of IP

law which are pretty well known yeah they are obvious but a lot of people don’t realize this and that’s a problem

because if you go and now right now you go on the street and begin asking people hey do you know who what do you know

about Vocahont what do you know about Mullen they’re going to say oh yeah that’s a Disney story it is not these

were actual people and their stories belong to a specific culture which

Disney has pretty much grabbed it has turn completely twisted copyrighted it

and then put it up for the masses in a way that has completely rewritten this

original story and most people if you go ahead and Google Pocahontas Mulan a lot

of the results are going to be about Disney and a lot of people just never knew and are probably never going to

know they actually existed and this is a form of cultural eraser and as a musician I also see this problem yeah

because if you hold on if you point this out to people I’ve seen this a million times you point out these examples of

absurd results of IP law then everyone who’s in favor of IP law they will say

something like well that’s an abuse of the system or that’s going too far or I’m not in favor of that but they never

have a principled opposition to the the very fact of IP which is the which is the c which is the reason for this

problem i mean you can point out a million examples and they will object to they will agree with you that every one

of those is ridiculous but they will say “But we still need protection from people’s ideas.”

So they they think you can have it both ways but you can’t have it both ways it’s the same thing with everything you know like they’re going to see the state

and they’re going to see oh well yeah the state has obviously killed a lot of people throughout history and maybe in

in just in the 20th century it killed over a 100 million people both like the

Nazis and the communists and even the United States and Europeans with

colonialism and they’re going to look at that and see that the state is like literally a mafia which puts a gun to

your head and forces you to pay them money just so that they give they can give you some sort of quasi public

service that is absolutely terrible compared to any private alternative and they’re going to go ahead and say “Oh

well yeah but we can reform it.” I agree it has never worked

but but there’s one difference the difference is this most people will admit that the the mist the pro the

problem with the state is that people have the mistaken idea that the state is necessary because without it you have

chaos so they’re mistaken about that but but at least they understand that the state is dangerous and that they most

people even most menus libertarians they regard the state as a necessary evil or

as an unfortunate an unfortunate use of force that we have to we have to accept

but they at least want to minimize it they realize that when the state murders people it’s wrong and so they try to

limit this with constitutional rights and limits but when you call intellectual property a property right

people think of it as a good thing and they see no principal reason to try to get so most people would say if we could

get rid of the state it would be a good thing it’s just impossible to get rid of the state we need the state unfortunately but for intellectual

property they view it as a property right that’s a good thing so in a way the the false belief in the in the

legitimacy and the necessity of intellectual property is a worst mistake because it’s more insidious and it’s

harder to fight well in that terms when whenever people go ahead and say well we

need intellectual property without intellectual property people are not going to be creating anything um so I

usually go ahead and say “Hey well if without intellectual property people wouldn’t have a reason to create

anything then we would probably not be having this conversation we would still be banging rocks living in a cave.” Like

do a lot of people seem to like think that oh yeah um everything we have now

was created because of intellectual property people had a reason to create things because they had a legal a legal

monopoly over their creations i’m not going to say that a lot of things were

created because there was an incentive because you could monopolize it but if

we look at history houses this is something you mentioned in against intellectual

property all the way back in 2008 but it’s a very good example um 2001 houses

go ahead yeah 2001 all right and then the version I I read was from 2008 anyway it was 2008 but that was based

upon a 2001 article all right um houses the first person who came out of a cave

and decided to build shelter by themselves they didn’t need any sort of

pent law to do that and what the thing is that people don’t create necessarily

because they’re seeking profits people a lot of time create because what they

think that whatever they they are creating is either going to have a cultural or like a recreational value to

them and to other people or it’s going to have a practical value to them and

throughout history many of the greatest inventions we can think of were made

without the existence of patent law or at least nothing existence of patent law

as we today know it because if we go back to in history

we know you you already had like copyright and patent law all the way back into like the 17th century mostly

in the UK um by the by the way it’s not just it’s not just inventions in the past it’s

also scientific discoveries right even now mathematical abstract scientific

discoveries theory theories of physics mathematical truths or ideas those are

not protected by by patents and yet we still have lots of I know but but the

point is in history you had lots of discoveries and inventions and innovation without patents and you had

lots of scientific discovery and knowledge develop without copyright or patent as well

yeah and but people seem to think that um everything we have now is the

creation of like it’s the byproduct of the existence of copyright law and pen

law and they usually think that for instance they many think that the

industrial revolution was partly or in great part thanks to the invention of

patent in copyright law all the way back in the late 18th century with stuff like

um I got rid of this but a lot of that was like um from the UK and that’s not

even the case and I mean if you go ahead and look at Venicia they had The

Venetian patent said stated Venetian patent um what what was it called venetian

uh yeah the Venetian statated I’m not really sure how it is pronounced which was passed in the 15th

century and according to for instance Iranian they would go well um surely

they would have passed that thing that protected um the creations of people and

gave them what basically was an early form of uh patents and Yeah random would say well

surely Venicia then became like a technological hub you know like the uh

the beacon of progress in the 15th century no it didn’t 100 years later

Benicia was completely in decay so there is no um historical evidence

like I don’t think there is any sort of actual proof to say that copyright and

patent actually promotes um innovation if anything it completely makes it

impossible to innovate because whenever you’re going to say “Hey I’ve got this great idea which I can modify this thing

that I have here and make it even better.” And once you do hey you cannot do that that’s our pent well you know

the the idea of u this kind of the early idea of of intellectual property and of

state granted monopolies is old i found um there’s one reference to this um this

cooking contest in ancient Cibbrus which is uh 20 2500 years ago and you know the

guy who won this this cooking competition would be awarded the right to be the only one who could make that

meal for a year um so this idea has been around a long time and it really

exploded after John Lock’s um idea that the source of rights is is

is your is your ownership of your labor and mixing it with something to be useful that’s kind of the essential

mistake that has permeated all of modern political philosophy is this idea that

creation of something of of value or use

uh or that can be sold is the source of property rights which is this which is the fundamental mistake

um behind modern justifications for intellectual property yeah um what you

mentioned there that that story about cyber it’s it is mentioned in the

DOist which is an ancient Greek text um I actually referenced that on my essay

and there which uh which which what’s the text no sophist or dnosophist I’m

not really sure how it is okay i’ll have to I’ll have to look that up i found it in uh another book uh but hold on just a

second i need hold on okay we’re back i’ll let you continue but let’s let’s try to wrap up in about

15 minutes so go ahead and get to whatever there’s just really just three more things that I want to speak about

and I think we can speak about this really quick the first is um as I was telling you before about cultural eraser

and all that um well I’m a musician i’ve written a lot of music i’ve recorded a

lot of music not necess It’s not a great deal of stuff i mean it’s certainly not

Pink Floyd or Dream Theater but um I’ve got my music and I come from a smaller

city and much of our music culture here is do it yourself so around here um I’ve

got a lot of CDs from local bands all of that stuff is not copyrighted it’s not

the um it has no photographic copyright over the songs and we’ve been working

very very much well in that sense and you know I think music is one of those examples of copyright not necessarily

being um a necessity you know like musicians

are not making music because they want to copyright it you’re

not going to a bar to a bar or to a restaurant to see a jazz band playing

their own music because they decided to write it so they could um register it and you know like sell it and I guess

it’s the same thing with you know like um writers and all that stuff you know all creative and recreational stuff it’s

not written because somebody wants to profit of off at least not in

principle um they turn off the light wait a

second okay so um one of the that what what I was saying is that

um people don’t create things you know like music or poetry or um books because

they want to profit out of it at least not in principle and that’s one of it depends maybe not monetary profit but in

the Austrian sense it’s for psychic profit but yeah well um when I mean profit it’s not monetary right um but

copyright is still exists in that sense and when you actually look at it then

you got a problem which is that copyright can actually lead to people not being able to create anymore or to

people not um being able to actually share their works later because when you

well maybe not you but let’s say a general example when you record a song

and you make it big and you record an album who ends up owning the phonographic rights and many times times

a copyright over that song is it you or the music label you signed to it’s

usually the music label it’s changing a little bit but usually in historically it’s been some publisher some some some

some publisher now what happens when you are somebody who has um let’s say political

views that are not very much in the mainstream you know like in 2025 cancel

culture is a thing and a lot of people have been driven out of the mainstream because of

their political views maybe you could we can argue all day long whether somebody

should or not or shouldn’t be driven out of the mainstream um through you know like social boycott and whatever i think

that that’s completely fine and that’s more of a thing that it doesn’t it is

not necessary to discuss because it’s voluntary in that sense but what happens when publishers own the copyright and

the rights over somebody’s work you you you can have them end up

pretty much censoring them yeah of course um I mean there was a I

don’t know if you there’s a new show called um the studio on I think it’s on Apple TV it’s about this guy that has a

music a movie studio there’s one little plot i think Seth Rogan’s in it and uh

there’s a plot where um um they buy the rights to a new movie

from Martin Scorsesei but then uh he it’s about

Jim’s Jones and Jonestown where they drink the Kool-Aid right but they’re also making a

movie about Kool-Aid and they don’t want to piss off Kool-Aid who gave them the copyright to use Kool-Aid as their movie

so they bought the rights to this movie and then they just killed it so that Martin Scarzes could not make a movie

about Jonestown because it would it would hurt their relationship with uh with with Kool-Aid so

there’s example now that’s f fictitious but things like that happen all the time where um if someone owns the copyright

to what you wrote they don’t have to publish it and they can stop you from publishing it i mean yeah that’s that’s

a problem and what I was saying is like you know um it it hasn’t happened yet

not necessarily like completely complete censorship of any person or not at least

not to my knowledge but you got musicians like Den Nugget or Moresy

who you are well known to be like rightwing and in the modern day they are

considered by many to be far right because for instance Stan Nugget is supportive of Donald Trump and their

music has been largely deemphasized by their labels because they don’t own it

so their labels are the ones who who are capable of judging whether their

music should be on Spotify whether their music should be promoted they don’t own

the copyright over their own music and there is like this very funny also this

very funny thing that happened way back in like 2020 which I mentioned in my

essay which is that Herman Lee the getist for a metal band called Dragon

Force he was a streaming and he would played one of the songs by his own band

and got DMCA strike like he got DMCA strike

for playing a song that he wrote yeah well of course it’s absurd i mean look I

I’m publishing an article in a book right now uh by a publisher and I wrote the article and they wanted me to sign

my copyright but I refused to do it because if I sign the copyright I can’t even use my own article later i can’t

republish it i can’t put it on my website i can’t uh make a ver make a new version of it i can’t translate it but

everyone else everyone else signs these things because they don’t care and they don’t understand but that is the

implication of this system that is true yeah it is sort of a double-edged sword i mean you can benefit a lot from

copyright as long as you can if you are able to completely keep the rights to yourself if they are yours but when they

are not yours you got you run into a problem and it’s one of these um self-reinforcing issues that um IP has

and it is that um this is something that I call the PPP doctrine uh patent

protect perpetuate PPP which is that copyright and patent law is it’s a

prisoner’s dilemma it exists right so if it exists

if somebody can’t protect this stuff then you’re going to want to protect your stuff not because you necessarily

believe in IP but because if you don’t then somebody else probably will and

you’re going to have your access to your own creation void because somebody else

that’s that’s a that’s a little bit confusing that’s more the case that’s more the case for patents than for

copyright um but but yeah go yeah I mean it’s most people have this false belief that if

you don’t copyright your music or your or your book well first of all it’s automatically copyrighted but they think

that you have to have a copyright in your work because otherwise someone else will will copyright it but only the

author of something can copyright it so that’s not really quite true and the same thing with with inventions if I

don’t patent my invention no one else can steal it because they’re not the inventor however if they independently

invent it and file a patent then they could stop me from using my own invention that is true yeah which is one

of the things that I had mentioned earlier about parallel thought um I was correct yeah i was being a perhaps a bit

reductionist for clarity sakes there but um well moving on from that though since

it’s better we wrap up um and the second and second to last thing I wanted to

mention is that something that I also mentioned in my essay is that patent

trolls and um not only patent trolls but the general issue with patents which is

that patent cost lives patent like I think that unironically patent if they don’t

kill people they prevent people from surviving mostly pharmaceutical patents

patents um worsen um patents worsen educational quality

particularly in underdeveloped countries like it is my case and pence this is

something that might sound a bit like um socialistic but I think patence is a way

that um they can be used to reinforce um the wealth of the more wealthier

countries ries which have historically let’s say um I want I don’t want to use the word exploited but which have

historically had an interest in um extracting resources from poorer

countries and I think that there is a 100 no you’re this is not anti- capitalistic it’s not socialist it’s

actually 100% true America in particular the west and America in

particular and in particular the pharmaceutical industry and Hollywood ollywood and the music industry are huge

beneficiaries they’re they’re recipients of wealth transfers from the rest of the world by means of patent and copyright

that is true yeah um India for instance India has a has a long history of

problems with biopiracy which is that almost entirely American people would go and find like the knee implant for

instance in India and say hey this is cool i’m going to patent it i’m going to patent like a and make me method of

extracting oil from this tree which has already existed for thousands of years

but they went ahead and made a patent out of it and then you s suddenly have a whole population of millions of people

who suddenly are infringing on the patent rights of that person who just

came upon this idea or this method or this um quasi creation of nature

and um they patented it like how how can many of these patents sure were later um

voided thank god but you know this happens and another problem is that

um patent trolls you know you got like these pent trolls who have historically

um patented rather menial inventions or random stuff you know like

in in big numbers and then suddenly when somebody has that idea independently and

tries to apply it to their own invention boom they are met with a lawsuit

absolutely correct and that’s something that I mentioned I mentioned various um

cases for in for instance NTP incorporated um which suit research in

motion because of some wireless email synology and then you got like floats

which suit a lot of um smaller AP developers all the way back in like 2018

2011 um because of like in application

purchasing mechanisms which is something that nowadays we got like in every single application on our

phones it’s like incredible that this kind of stuff exists and all right

intellectual ventures pretty sure you know that name but it’s like one of the greatest patent trolls in history like

those guys own a metric [ __ ] ton of patents and whenever somebody infringes

on that usually from parallel to thought or just because they are really good

inventions that um need to be used they go ahead and file a lawsuit or naturally

they can um force them to pay a fee to use their patent and then you got like the

pharmaceutical um industry with its patents and it’s evergreen and in the case of the US you

know you got the insulin market which is is insulin in the US is a stupidly

expensive and it is in great part because of um evergreening you know like

the patents on insulin are all the time um renewed

because they just make a small adjustment in the formulas of the insulin or in the way it is administered

and boom you get like a 10year extension on the patent of it so that delays a lot

the um the possibility for how do you call it um not copies

um bio generic mean bioimilars yes generics and bioimilars from entering

the market so then there is like this one case with

COVID COVID vaccine um which back in like back in like

2021 the UK was um the UK government was

blocking attempts to wave the intellectual property over some of the vaccines and treatments yes

while it is estimated that in average some 10,000 people were dying every day

because they didn’t have access to the vaccine in great port because of that um

patent problem and then again this really has a huge effect on poorer

populations poorer countries because of course if you live in say Bolivia or

here in Argentina too but maybe Argentina is a bit more developed than that but if you live in a poor

country um you cannot really have like a national company a local company which

produces um copy patented medicine because you know

patents apply globally thanks to you know like the WTO and the WO no patents

are patents are national and usually what happens is when you get a patent it’s in one country but usually you file

similar patents in other countries but it’s too expensive to file it in every country so what you do is you file it in

enough countries where you can you can block the countries that produce it um

so for example if if if a given drug is made in five countries like the European Union and Canada and America or whatever

or Japan China you you file patents in those countries but then there’s nowhere

that they can’t sell it to another country like Brazil or whatever um even if you don’t have a patent there because

there’s no one that makes it but but go ahead that’s roughly Well thanks for correcting me on that but um I mean even

I think that even then I mean if you I were to open up like a um like a company

here that produces some drug that is patented in the US but not here i’m pretty sure that eventually the patent

holders would come around knocking and would file a patent for this country

well if well what they would do is if it’s too late they would they would they would suffer the loss there but um uh

but then the the Brazil industry or the Argentinian industry could not they

wouldn’t emerge in the first place because they wouldn’t be able to sell their drugs in other countries um uh

like in the big countries of the west u so it’s unlikely that you would happen and it if if Argentina started becoming

a pharmaceutical hub then then that’s another country that the big companies would file their patents in so yeah they

would they would find a way to whack the mole and to stop it there that’s true so yeah not only so not only is

pharmaceutical uh stuff a problem but then also as I was saying education and this relates more to copyright and it is

like I’m not sure about you you had your education well a long time ago but now

um myself um I went all my life to public school public university too

and copyright means that you cannot copy something that is

copyrighted and you cannot pirate a book that is copyrighted but guess what um we

Argentines are not really um we don’t have the level of wealth that Europeans

or Americans have so we cannot go around affording expensive books which costs um

well I’m not going to say they cost a massive amount but which have a

impactful cost for especially for middle class people and lower middle class people

so you don’t go around buying those books so what we did in school was the teacher would usually make us make um

photocopies of all these books and we’d use those photo copies of course to study and even when we got to university

I was reading books about Python and network and all which were pirated

online because the teachers told us yeah um here’s the link pirateed like the

teachers you know and you think that I mean in US this would probably if this

was done like at Harvard then that would probably end up in a lot of um what

happens is yeah the publishers they turn a blind eye to massive copyright in the third world because they go after the

the lowhanging fruit and another thing that sometimes happens is you’ll have a publisher like say wy there’s a famous

case so you’ll have these publishers and they will publish a cheap version on cheap paper like like let’s say they

sell a textbook for $200 in the US they might sell it in India for $15 or or or

$30 but then what happens is you have arbitrage happen so you have

someone they go to India or Argentina or whatever and they

buy those copies and they bring them back to the US and try to sell them there and of course there was a big

lawsuit wy it’s a famous lawsuit wy tried to sue and say that if you import a copyrighted work um it’s still

copyright infringement uh under this thing called the exhaustion doctrine and then there was a big legal case where

the court I think the court said um I think they sided with um against the publisher i could be wrong but it had it

was a big legal battle first but that’s just an example of how they will do everything they can to maximize their

profit and it will it will it will hurt people it it does hurt people and well it’s the same with research papers you

know like a lot of research papers are naturally copyrighted and you need to pay a lot of money to access that which

also means that people in porter nations have like a completely limited access to

a lot of knowledge and a lot of tools to aid their own um progress which is

incredible when you think about it and you know may maybe we are not yet there

but we are quickly moving into a world where surveillance and especially the way that copyright is

um well this may be before your time but you know the guy that this guy um u the

young guy Aaron Schwarz who was the co-inventor of RSS and podcast you know some technology um he was a student and

he took a laptop and used the I think Colombia one of the one of the

university servers and he uploaded thousands of these research papers to the internet which was a copyright

violation and he got um charged with copyright criminal copyright violation

he was this is just a 20 23 year old guy and he was threatened with

like dozens of years maybe life in federal prison so he committed suicide

he he he killed himself because um he was going to be basically killed because

of copyright um that’s just one example of many that’s one example is um the

creator of FM radio technology Edwin Howard Hamstrom he had created FM radio

technology the AM industry opposed him rca copied

his patent without any license yeah he had to go to really extensive

um uh legal battles over that which completely ruined his psyche he ended up

like having breakdowns and paranoia his wife tried to commit suicide I believe

and he ended up committing suicide himself because of all of all this stuff

which is completely like it’s mind-blowing so just to um be done with this like the

last thing we can talk about and probably we’re going to get over this over with um rather quickly is like the

last part of my essay deals with not

abolition but with reform and this is not because I am against abolition i

mean if it was up to me I’d completely get rid of copyright and patent and

trademarks and everything tomorrow but it’s not um realistically

real realistically viable not yet so I dedicated this part um with a what I

consider to be a more realistic solution to what’s an absort problem which is well IP

and just to go over it a little bit my

um my main proposal to speak up a little bit you’re you’re you’re getting quiet yeah um my sore is kind of uh my my

throat is kind of sore now um but my main proposal is to replace uh

exclusivity IP exclusivity with what’s basically

a I would call it a royalty system with

um decaying exclusivity so the idea the idea is instead of copyright and patent

law saying hey you cannot use this copyright and patent law should grant people shorter

terms and they should make it so when

the patent comes into effect or the copyright comes into effect um they h

they would have a right over 100% of the profits or the income of any work or any

creation or the just the general um usage and marketing of their own

creation and this would be decaying over time this rate would be decaying over

time until it reaches zero which is when basically the patent would expire so let’s say that I make a new thing like

let’s say it’s a new storage device i patented it and so say the patent lasts

um five years so for you divide um those five

years by um you divide yeah those five years by 100 or so and you got um that

declining every every certain amount of time the amount of royalties you can

claim over what other pe over the use of other pe uh over other people’s use of

your creation it’s going to be that number right so it first is 100% then a

certain amount of time passes and it’s 99 then 98 then 97 to 96 so what happens

here is that other people can use your creation they can market it they can do

whatever they want with it but they have to they have a legal oblig obligation to

pay you they have to pay you a percentage i I I recently had a a podcast about this

a couple days ago it’s it’s there’s lots of proposals to reduce the cost of IP

law and one of them would be for patents you would get rid of the um well you

could change the term you could also uh you could also get rid of the injunction in other words all you can do

is demand a payment um that would that that would definitely reduce the harm of

IP law i don’t know why it’s practical because everyone that supports IP would oppose that just like they oppose it now

well sure think about it it’s actually per perhaps a much a much more

egalitarian um system because a lot of people defend

um IP on a solitarian grounds but if you think about it um if you allow people to

use patented stuff while also paying the creator a

fee and also giving a ex a not exclusivity but um advantage to the

creator by giving them the capacity to create their um to produce their patent

and to produce their creation at a rate that’s way below below that of

their competitors because their competitors would have to pay royalties then you could have a market

that’s not monopolized and while sure you would have like the this one company this one

organization or one individual who would offer the same product at the cheapest

price you could still have people be willing to pay more for an alternative

that might be of a better quality or have um something in a specific that they like which is if you think about it

it’s the best thing you can possibly have which is much better than just exclusivity and monopolizing any

creation it’s not it’s not the patent people they believe it’s a property right okay so they they look at it like

you have a you have a you have a an office building and right now you can physically prevent someone from using it

without your permission and then you can use that right to exclude them to say if you want to use my building you have to

pay me whatever I whatever we negotiate but if you take away that right to stop

them that reduces part of your bargaining power this is exactly why the IP and the patent proponents in America

go crazy when the Supreme Court law makes it harder to get an injunction um

you’re right that that would make it better because it would reduce the harm but it would reduce the harm by reducing

the strength of the patent right and everyone’s going to oppose that just like they oppose getting rid of patents

or reducing the term so all you’re doing is produ you’re you’re coming up with ways that you could reduce the harm of

patents or IP and you’re saying it doesn’t hurt

that much it still gets you some of what you want that is true but they’re still going to want everything they can get

the IP because they because I don’t think they really um defend IP because

on authoritarian grounds or ethical grounds but they defend IP because they

have just a preconception that is deeply ingrained into them and it’s the same people I guess it’s the same reason why

a lot of people still defend um Marxism when it’s been proven that well it

doesn’t work in practice it’s it’s a complete mess and you got all of history

to prove that but a lot of people still offend that so in that sense I guess

this is not really something that you would talk about to someone who is completely um in line and happy with the

current state of IP but this is something that you wouldn’t have to bring up to somebody who is in the

middle or who is not very well informed on that like you explain to them the problems with IP and you say and you say

hey well we don’t have to abolish it because a lot of people are going to are going to hear well abolition that’s

that’s crazy what is wrong with you but if you tell them well we can reform it then you can um convince them to support

that idea and I think that if you No you can’t no you can’t no you can’t I’ve

tried you can’t I well I’m going to try myself and go ahead maybe maybe I’m

going tell you someday um but I think that if you we if you are if you are capable if we are capable of convincing

people that to reform IP in a way that reduces its harm and um reduces its grip

on on the market of ideas we could have abolitionism

um become more mainstream in the future you know I it’s my personal opinion like

if you go for it go for it if you Yeah i mean I think I think that we need to shift the overtone window on this um

well you know technology but the thing is technology and reality is shifting the overton window because most people

like say in third world countries like you say they don’t they copy and they

pirate and even in the west students and regular people they they pirate all the

time but companies can’t do it companies can’t do it you know so that’s the

problem is you the big companies still have to obey the law because they could be easily sued but yeah your average

person doesn’t think there’s anything wrong with um with piracy on a small scale and they do it all the time even

people that are procopyright I guarantee all these IP people these Irandians and these

menarchists these utilitarians when they’re doing their research and they’re browsing things on the internet

um they sometimes go get a pirated copy of a book because it’s easier and there’s no other way they do it all the

time so technology is slowly reducing the I mean you remember these commercials in America anyway anyway the

RAIAA put out where they say you wouldn’t you wouldn’t download a car so

why would you download a song and people are like I would download a car if I could download a car I would download

you can 3D print a car now exactly you can 3D print a car you when you 3D

printed maybe you are you’re infringing on copyright but hey you can pirate a literal car in 2025 five and um a lot of

people have this idea you know like like one of the things that I hate that is that a lot of people think that oh well

people people will keep pirating i mean people pirate because they like to pirate um people pirate because they

don’t like to pay and maybe that’s true but if people were to pirate because

they don’t want to pay just because they don’t want to pay then I mean what I tried to get to

here is that hey copyright is not is stopping people from pirating no it’s not stopping them it’s not stopping

people from pirating and this is something that appears on your website and it is that um we are not supposed to

support a failed um business model and what is a failed business model one that relies too much

on copyright and that it relies too much on the state enforcing its

existence do are you familiar with um a platform like Steam

well yeah steam what does Steam do um Steam sells video games a lot of video

games thousands of video games and it also sells the soundtracks for video games and you even it sells movies and

some hardware and software and all and Steam is widely used like people use

Steam they’re I think there is well over a billion users by

now why people why do people buy games on Steam when they can just go ahead and torrent them i mean it’s more it’s more

convenient it’s Yeah it is more convenient and some people would say “Hey well a Steam exists because copyright exists if copyright didn’t

exist then people wouldn’t buy games on Steam because Steam wouldn’t have a reason to sell them well they they would

people no they would they might it might be cheaper but no they you still have a service like Steam people buy games from

Steam because first of all it’s much safer than going into these Russian

websites where anything you click is probably going to download a Trojan to your computer um they get automatic

updates on their games they have a whole integrated ecosystem of you know of

trading cards and inventory digital inventory items which are kind of like

NFTTS and in um an internal market in Steam

like you literally have an internal market in the Steam people buy and sell things based on offer and demand which

is something incredible people buy games from Steam because it is convenient and it is a

good and e steam is a good business model because it is convenient it

provides people with something that they always wanted like somebody in who was

buying and playing games in 2000 had to go ahead and go to a GameStop or some

other place find a game and judge that game by the cover art go home put a put

on a CD install it put on the CD to play the game they wouldn’t get updates it

was like not convenient at all it was a pain in the butt for anybody who liked

to play games and now 2025 you can just go ahead turn on your computer go to

Steam find any game you want you can look at hundreds of pictures and videos

and game plays and streams about it and read people’s reviews and you just click

three buttons you bought it you can download it and you can play it and if there’s an

update you’re going to get it automatically you can unlock achievements which actually achievements

are not meaningful at all real but they just feel good to unlock it just feels

good to have like that notification appear on your screen that says “Oh you killed 100 enemies congratulations you

got this achievement.” And that’s that’s just a good business model and it

doesn’t need copyright to exist it doesn’t and it’s the same with platforms like Netflix or

Spotify like a streaming services and all these things are just a living proof that copyright is not necessary to stop

piracy it is a proof that piracy is stopped by making things more convenient

by making people prefer to pay instead of having to pirate and even I myself

have engaged in buying stuff even when I can’t pirate it of course lots of people do by the way we need to wrap it up but

let me let me mention I think you should go do what you want to do try to do everything you can if you come up with good ways to persuade people especially

in Argentina that’s good if you can persuade me to do more that would be good um um by the way the title of your

paper is almost the same as one of my papers which is called ideas are free um

you can use it i don’t care but my my my paper is called ideas are free the case against intellectual property law it’s

almost same as your title yeah well it’s um it’s something that I mentioned in the end it’s that these ideas um like

everything I wrote is not my idea but it’s the ideas of like all the people who have shared these things me too

that’s how ideas work yeah but um I got I’ve got to wrap it up i enjoy talking

to you i want you to Good luck with everything feel free to contact me again but um one thing that I I just want to

recommend you a book if you have time to read it later um it’s um the way of Changu by Thomas Merton it’s a book on

Tao philosophy but even if it’s like thousands years old um trust me it has a

lot of libertarian values in there like proto libertarian values I’m pretty sure you’re going to like it burton M- E R T

O N uh yeah all right i’ll look it up there’s a I’m going to there’s a book

the book I’ll put on my show notes it’s um it’s called it’s a it’s a uh it’s called uh

uh it’s something about an ancient Chinese uh saying uh to steal to steal a

book is an elegant offense and it it shows that in some cultures they don’t

regard copying ideas as an insult or as a harm it’s sort of a compliment you

know um yeah well um for thank you very much for this i never expected to have a

conversation with you so this was very much enlightening all right well good luck with everything

and we we’ll we’ll stay in touch have a have a nice weekend see you hey take care bye bye

Share

© 2012-2025 StephanKinsella.com CC0 To the extent possible under law, Stephan Kinsella has waived all copyright and related or neighboring rights to material on this Site, unless indicated otherwise. In the event the CC0 license is unenforceable a  Creative Commons License Creative Commons Attribution 3.0 License is hereby granted.

-- Copyright notice by Blog Copyright