Author(s): Ratna Gopal Singh and Khushi Yadav
Paper Details: Volume 3, Issue 4
Citation: IJLSSS 3(4) 09
Page No: 76 – 88
ABSTRACT
The rapid proliferation of Artificial Intelligence (AI) has fundamentally disrupted conventional notions of creativity and innovation—concepts that are at the heart of intellectual property law. In India, the Copyright Act, 1957 and the Patents Act, 1970 were drafted at a time when the idea of non-human authorship or inventorship was inconceivable. These laws still hinge on the assumption that an author or inventor must be a natural person. However, with generative AI systems now capable of producing music, literature, art, and even new inventions, the existing statutory frameworks are being stretched beyond their intended limits.
This article investigates the core legal dilemmas posed by AI-generated works and inventions, particularly in terms of authorship, ownership, originality, and inventiveness. It delves into landmark Indian judgments such as Eastern Book Company v. D.B. Modak [(2008) 1 SCC 1] and Bishwanath Prasad Radhey Shyam v. Hindustan Metal Industries [(1979) 2 SCC 511] to trace how Indian courts have interpreted the requirements of creativity and novelty in the past, and how these standards might be challenged in the AI era.
Keywords
Artificial Intelligence, Copyright Law, Patent Law, Indian IP Legislation, AI-generated Works
INTRODUCTION: AI IN THE AGE OF CREATIVITY AND INVENTION
Something strange is happening in the world of law—something that’s making lawyers, judges, regulators, and lawmakers scratch their heads. The age of Artificial Intelligence isn’t coming anymore; it’s already here. From AI-generated paintings to algorithmically designed inventions, we are witnessing a silent revolution that’s disrupting the most human of all legal ideas: intellectual property. What was once considered the exclusive preserve of human creativity and ingenuity is now being tested by non-human creators, i.e. machines that can write poems, compose music, design blueprints, and even suggest novel chemical compounds. All of this raises a fundamental question that our existing laws are unprepared to answer: who owns what when the creator is not human?
Patent and copyright law in India, as elsewhere, rest on the idea of rewarding originality and invention. These laws are, at their core, meant to protect the output of the human mind. But when AI can create music indistinguishable from a composer’s symphony, or design mechanical structures no engineer has thought of, the law begins to stutter. In its current form, Indian IP law—like most others globally—has no proper answer to the rise of AI-generated content. And what’s more worrying is the silence in our legislative corridors. While other countries have at least started discussing frameworks to deal with AI authorship and inventorship, India’s laws still operate under the safe assumption that only human beings are capable of thought, intention, and creativity.
But this assumption is beginning to crack. In just the last few years, we’ve seen legal battles emerge across the world: copyright offices refusing AI-generated artwork protection, patent offices rejecting applications listing AI as the inventor, and courts stepping in to interpret legal definitions that never considered machines. As a law student trying to understand this moment in history, it feels like we are witnessing a constitutional moment in IP law. The values that underpin our understanding of authorship, originality, novelty, and creativity are being forced to evolve. And yet, even as the world is forced to confront this change, Indian law remains trapped in older paradigms. Through this article, we’ll try to explore how AI challenges our current framework, how Indian courts have historically interpreted human creativity, and why the time has come to enact bold, clear legislation to deal with the growing overlap between AI and intellectual property.
THE HUMAN TOUCH: HOW COPYRIGHT AND PATENT LAW GREW AROUND PEOPLE
Before we dive into AI, it’s important to understand where our copyright and patent laws come from. Both systems are built around a very human idea—that people who think, imagine, and create something new deserve protection. The Copyright Act, 1957, although repeatedly amended, still assumes that the ‘author’ is a person. In the same way, the Patents Act, 1970 speaks of the ‘true and first inventor’, without imagining that a machine could ever play that role. This isn’t a coincidence. Both laws are rooted in the notion of reward for intellectual labour. Copyright protects the original expression of ideas, while patent law protects new and useful inventions. In both cases, the law assumes that these ideas and inventions are the product of human mental effort, shaped by emotion, reason, and subjective experience.
Historically, courts too have echoed this idea. In the realm of copyright, the landmark judgment in Eastern Book Company v. D.B. Modak (2008) 1 SCC 1 clarified that originality requires a minimum degree of creativity not just labour. This idea of a “modicum of creativity” has since become central to copyright jurisprudence. The court moved away from the “sweat of the brow” doctrine, which had earlier allowed protection merely for effort, and instead recognised that copyright requires something more an intellectual contribution that reflects individuality. In short, it must be human.
Similarly, in patent law, the notion of inventorship has always been tightly tied to human creativity. The requirement of “inventive step” under Section 2(1)(ja) of the Patents Act implies an act of mental effort that goes beyond obvious application. The jurisprudence in Bishwanath Prasad Radhey Shyam v. Hindustan Metal Industries (1979) 2 SCC 511 reaffirms this idea by emphasizing that an invention must reflect ingenuity and skill. Nowhere in these discussions did the court anticipate that a machine might one day meet, or even exceed, that standard.
so when AI steps in and performs what appears to be original creative acts, the whole logic of our IP system begins to wobble. After all, how can you apply tests like “creativity”, “originality”, or “inventive step” to an algorithm that is merely executing programmed instructions even if the results look stunningly innovative?
AI AND COPYRIGHT: CAN MACHINES MAKE ORIGINAL WORKS?
Let’s now step into the thick of the copyright dilemma. The moment an AI model like DALL·E creates a beautiful painting or ChatGPT writes a compelling poem, the question arises who is the author? If we look at the Indian Copyright Act, 1957, particularly Section 2(d), it defines an “author” in a very person-centric manner. For example, in the case of a literary or dramatic work, the author is the person who creates it. In the case of a photograph, it is the person who clicks it. In every form, the law refers to a “person” and not an entity, and certainly not a machine. So, when an AI tool creates something independently without direct line-by-line human input should it be eligible for protection? Or do such works fall into the public domain by default?
Globally, the consensus is still firmly on the side of the human. The U.S. Copyright Office, for instance, in a 2023 case involving an AI-generated graphic novel titled Zarya of the Dawn, clearly held that the images generated by AI were not eligible for copyright protection. Only the parts of the work where the human had intervened like structuring the storyline or selecting the arrangement were protectable. Closer to home, India hasn’t yet had a landmark judicial ruling on AI authorship, but the legislative language and interpretive traditions strongly indicate a similar approach. The requirement of a “modicum of creativity”, as established by the Supreme Court in Eastern Book Company v. D.B. Modak (2008) 1 SCC 1, suggests a subjective human effort, not algorithmic computation. Even in R.G. Anand v. Delux Films (1978) 4 SCC 118, the Court emphasized the role of human imagination and judgment in artistic expression, noting that “ideas are not copyrightable, only their expression is” and that expression must reflect the author’s personality.
But here’s the twist. While AI may not qualify as an author under the current law, the outputs it creates are increasingly indistinguishable from human work. This creates a loophole of sorts. What happens when someone uses an AI tool to generate an entire album or writes a novel by carefully curating AI-generated responses? If the human only enters prompts but does not write any actual words or compose any music, do they still count as the author? Should we view them like a director who orchestrates a performance, or are they more like a technician operating a machine?
This grey area where human input and machine output blur is where the real danger lies. Without proper legal recognition, AI-generated works risk being excluded from protection entirely, which might encourage plagiarism, unauthorized reproduction, and misuse. On the other hand, granting full copyright to such works might flood the copyright registry with millions of instantly generated outputs, diluting the value of genuine human creativity. There’s also the issue of training data many of our AI models have been trained on copyrighted material, including books, songs, films, and academic articles. If the output of such a system is partly derived from that training data, does it constitute infringement? In the absence of clear legal guidance, courts may be forced to interpret outdated statutes through a 21st-century lens a difficult and risky exercise.
JUDICIAL INTERPRETATION AND THE SUPREME COURT’S TAKE ON CREATIVITY
So far, Indian courts have been fairly conservative but also principled in their interpretation of creativity. As mentioned earlier, Eastern Book Company v. D.B. Modak was a turning point. The court rejected the idea that mechanical or repetitive labour deserved copyright protection, insisting on the need for creative spark. This is significant in the AI context because it draws a distinction between mere effort and genuine creativity. A machine, no matter how complex, cannot have intention, emotion, or perspective. It can mimic, but it cannot originate. Even if AI-generated content looks original, it is arguably the result of statistical pattern recognition, not creative thought.
In another important case, University of Delhi v. Kamal Singh (2015) 13 SCC 651, the Supreme Court again reiterated the human-centric approach to authorship and originality. While deciding a dispute over academic material and course content, the Court noted that academic works involve “intellectual contribution, analytical thought, and personal reasoning” none of which could be attributed to automated compilation. These judgments, although not directly about AI, provide useful insights. The Court has consistently treated originality as a reflection of human personality and not as a mechanical process. That doctrine sits uncomfortably with AI, which lacks personhood, consciousness, and subjective choice.
Similarly, Indian Performing Right Society Ltd. v. Sanjay Dalia (2015) 10 SCC 161, though more about territorial jurisdiction in copyright disputes, reaffirms that the law aims to protect and reward “authors and creators” who have contributed something unique to the cultural domain. The entire tone of the Court’s analysis reinforces the centrality of human agency in intellectual property disputes. Taken together, these rulings form a judicial philosophy a jurisprudence of human creativity that may make it difficult for Indian courts to suddenly accept machine-generated works as eligible for protection.
Yet, there is an undercurrent of openness in some of these cases. Courts have often acknowledged the need to interpret laws in light of changing technologies. In Justice K.S. Puttaswamy v. Union of India (2017) 10 SCC 1, the Supreme Court famously declared that the Constitution is a living document and must be read in tune with technological and societal developments. While that was a privacy case, the sentiment applies equally here. The question is, will Indian courts be willing to extend their flexible interpretive approach to copyright and patent law, or will they insist on legislative clarity before making such a leap?
AI AND PATENT LAW: INVENTORS WITHOUT MINDS?
Now if copyright law is struggling with the question of authorship, patent law is in a full-blown identity crisis. The notion of inventorship is foundational to the patent regime, and it’s hardwired into the Patents Act, 1970. The law consistently assumes that an inventor is a human being who applies mental skill and insight to solve a technical problem. But what happens when that inventor isn’t a person at all, but an AI algorithm trained to generate novel solutions, products, or chemical compounds? Can such a machine be credited as the “true and first inventor”? And if not, then can any patent be granted at all for the invention?
Section 6 of the Patents Act is clear that only a person can apply for a patent. The applicant must either be the true and first inventor or someone claiming through him. The key word here is “him” the assumption of personhood is baked into the statute. Section 2(1)(y) defines “patentee” as the person for the time being entered on the register as the grantee or proprietor. The entire scheme of the Act assumes that an inventor has legal personality, capacity, and agency. Machines, at least in Indian law, have none of these.
This has led to interesting developments worldwide. Perhaps the most famous is the saga of DABUS the AI system developed by Dr. Stephen Thaler which autonomously generated two novel inventions. Dr. Thaler filed patent applications in several jurisdictions listing DABUS as the inventor. The UKIPO, the EPO, and the USPTO all rejected the applications on the grounds that inventorship must reside in a natural person. In India, there hasn’t yet been a DABUS-style test case, but if there were, the result would likely be the same. The logic from cases like Bishwanath Prasad Radhey Shyam v. Hindustan Metal Industries (1979) 2 SCC 511 still governs our approach. The Court in that case stressed that an invention requires human ingenuity and not just novelty or utility.
But here’s the real challenge: AI is already capable of inventing things that are novel, non-obvious, and industrially applicable, the three pillars of patentability under Section 2(1)(j) of the Act. In fields like drug discovery, materials engineering, and electronics, AI systems are routinely generating ideas that human researchers might not stumble upon for decades. If such inventions are left unpatented merely because the inventor lacks human consciousness, we may create a dangerous disincentive for innovation. Companies might simply hoard such technologies as trade secrets, avoiding disclosure altogether. This would run contrary to the public purpose of patent law, which is to strike a bargain between innovation and dissemination.
There’s also a growing tension between form and substance. If a human simply tweaks or finalizes an AI-generated invention, can they claim to be the true and first inventor? That feels legally weak and ethically dishonest. And yet, unless the law is amended to either recognise AI as an inventor or acknowledge collaborative inventorship between AI and humans, this may become the norm. This grey zone will only expand as AI systems become more autonomous and more creative. Today, they might need prompts and inputs; tomorrow, they may need nothing but a problem statement and access to data.
One possible way forward is to allow human entities to claim inventorship while making a statutory declaration of the role played by AI in the process. But even this would require a reworking of the existing statutory language. As of now, the law simply doesn’t anticipate non-human inventors. That’s not just a theoretical problem it’s a gap that could derail innovation and leave Indian companies at a disadvantage compared to jurisdictions that adapt faster.
THE EVOLUTION OF THE SUPREME COURT’S VIEW: INDIAN CASE LAW AND IP PHILOSOPHY
Indian jurisprudence on intellectual property has always placed strong emphasis on human-centric standards of creativity and inventiveness. But over the years, as technology has evolved, so too has the language of the Supreme Court. While laws have remained largely unchanged in form, the Supreme Court of India Judicial interpretations have slowly digress to reveal an openness toward complexities of the upcoming technology, even if not specifically toward AI. This evolving judicial mindset could be crucial in navigating the current disruptions caused by artificial intelligence in copyright and patent frameworks.
Let’s begin with copyright. In Eastern Book Company v. D.B. Modak (2008) 1 SCC 1, the Court took a significant step forward by discarding the older English “sweat of the brow” doctrine and endorsing the standard of creativity that required “some degree of creativity”. This wasn’t just a cosmetic shift; it changed the tone of copyright protection by asking whether the author had exercised independent judgment. The decision placed India closer to the U.S. “modicum of creativity” test laid down in Feist Publications, Inc. v. Rural Telephone Service Co., a case often cited globally for setting a creativity threshold. But the important takeaway here is that the Court recognized that law must reward intellectual choice—not just effort. This creates a problem for AI-generated works which, by their very nature, operate without cognitive faculties.
In patent law too, a notable example is Bishwanath Prasad Radhey Shyam v. Hindustan Metal Industries (1979) 2 SCC 511. Here, the Court elaborated on what counts as “inventive step” and emphasized that patents are not to be granted for trivial improvements or mere workshop improvements. The Court was clear that a patent must reflect real ingenuity. Again, this highlights a subjective standard a flash of human creativity that might be missing in AI-driven innovation. What’s interesting, though, is that both these judgments are concerned less with who creates and more with how the creation reflects originality or inventiveness. That subtlety might be useful going forward.
A more modern illustration is the case of Tata Press Ltd. v. Mahanagar Telephone Nigam Ltd. (1995) 5 SCC 139, where the Court recognised that technological change requires dynamic interpretation of IP laws. Though not about AI, the Court in that case accepted that what was once considered non-patentable may need reconsideration due to changing times. It’s a recognition that intellectual property law isn’t static it breathes, and it must keep up with science and society. This sentiment echoes even more loudly in Indian Performing Right Society Ltd. v. Sanjay Dalia (2015) 10 SCC 161, where the Court acknowledged the impact of digital platforms on copyright law, hinting that the traditional territorial and temporal rules may no longer serve justice in an online world.
But perhaps the strongest example of this jurisprudential evolution is Justice K.S. Puttaswamy (Retd.) v. Union of India (2017) 10 SCC 1. Though primarily a case about privacy, the Court’s reasoning about technology is worth noting. It held that constitutional rights must evolve to protect the individual against new forms of technological invasion. If constitutional interpretation can be so future-facing, why not statutory interpretation in IP law? The real message here is that the supreme court when pressed, have shown themselves willing to reimagine old legal concepts in light of modern challenges.
That said, this flexibility also comes with limits. Courts can only interpret the law; they cannot rewrite it. In the absence of clear statutory amendments, the judiciary remains tied to the text of the law, however expansive its spirit might be. That’s where the need for legislative intervention becomes urgent a subject we’ll explore next
THE NEED FOR STRINGENT LEGISLATION TO BALANCE AI AND IP PROTECTION
At this point, one thing becomes painfully clear our existing intellectual property laws are outpaced by technology. While courts have stretched interpretations to accommodate modern realities, there’s only so much that judicial elasticity can achieve. To effectively manage the disruptive impact of AI on both patent and copyright law, we need targeted, robust, and forward-looking legislation that can provide clarity where current law is either silent or vague.
Let’s begin with the problem of uncertainty. As we come across it today, neither the Copyright Act, 1957 nor the Patents Act, 1970 even contemplates the possibility that a non-human entity could be an originator of intellectual property. This vacuum creates not only confusion for rights-holders but also a playground for exploitation. If someone uses an AI model to generate thousands of works or inventions, and then claims human authorship merely because they pressed a few buttons, it dilutes the spirit of IP law. At the same time, denying such works any protection creates a risk of unregulated exploitation, piracy, and unfair commercial advantage. A balanced statute should neither romanticise AI as an artist nor demonise it as a threat. It must recognise the nuanced role of AI as a tool that may sometimes cross into the terrain of co-creation or co-invention.
One possible route is to legislate a middle path wherein AI-generated works are protected but attributed to the person who “arranged for” the AI to create the work, much like how a producer is credited in the film industry. This approach, sometimes called the “AI-assisted authorship” model, allows for human accountability while not entirely erasing the contribution of the AI. For patents, the law could consider creating a new category of “algorithm-generated inventions” where the AI’s role is declared, and the human controller or institution behind the AI gets the rights. This isn’t too far-fetched; legal systems have long recognised derivative liability and indirect authorship in other contexts.
But these aren’t just theoretical fixes. Legislative action is essential for legal certainty. For instance, the U.K.’s Copyright, Designs and Patents Act 1988 already contains a provision in Section 9(3) that attributes authorship of the computer-generated works to “the person by whom the fixtures made necessary for the creation of the work are undertaken.” A similar model could work in India, but it would need to be carefully tailored. We’d have to define what kind of “arrangements” qualify, who holds moral rights, and whether duration of copyright should differ for AI-generated works.
Then there’s the issue of transparency and data provenance. AI models are trained on massive datasets, often scraped from the internet without consent. This includes copyrighted books, music, films, articles works that are protected and monetised. Current law offers little guidance on whether such training constitutes “fair use” or infringement. A new legislation must set clear rules: What kind of data can be used to train AI? Should consent or licensing be mandatory? How do we measure compensation for the rights-holders whose works fuel these machines? This stays not just as a legal questions, but also ethical and economic ones.we risk eroding creators’ rights in the name of progress.
Equally critical is the need for an independent regulatory body that can monitor, guide, and enforce AI-related IP practices. Think of it as a “Censor Board meets IP Office”—an autonomous authority that not only registers AI-generated works and audits training datasets, but also offers advisory opinions on disputes. Given the speed at which AI is evolving, courts alone cannot handle the burden of resolving every new ambiguity. A dedicated regulatory framework can act swiftly, establish standard practices, and help prevent litigation overload.
Finally, the legislation must address enforcement. What remedies are available if someone uses an AI tool to generate a near-identical copy of your copyrighted song? How do you even prove infringement when AI outputs are probabilistic and difficult to trace? The new law must incorporate digital tracking tools, watermarking protocols, and reverse-audit systems that help creators trace AI-generated duplicates. Otherwise, we may end up in a legal battlefield where the weapon is invisible and the victim is untraceable.
COMPARATIVE GLOBAL APPROACHES: LESSONS FROM AROUND THE WORLD
While India is still at the stage of judicial speculation and legislative silence, several other jurisdictions have already begun grappling—albeit imperfectly—with the legal implications of artificial intelligence on intellectual property. This global perspective offers us valuable lessons, particularly as we consider possible reforms to our own laws.
Let’s begin with the United Kingdom. As mentioned earlier, the UK Copyright, Designs and Patents Act, 1988 already includes Section 9(3), which directly addresses computer-generated works. It says that for such works, “the author shall be taken to be the person by whom the arrangements necessary for the creation of the work are undertaken.” This provision has proven prescient, though it was drafted well before the age of generative AI. The UK’s approach is refreshingly practical it acknowledges that AI lacks legal personhood but ensures that someone, usually the programmer or commissioner, can be the beneficiary of the work. However, the scope of “arrangement” remains undefined and judicial interpretation has been minimal, making the provision more of a placeholder than a solution. Still, it reflects legislative intent to adapt copyright law to automation.
The United States has been less accommodating. The U.S. Copyright Office has taken a firm stance that only works created by human authors are protectable under the Copyright Act. In 2022, they denied protection to a comic book created using Midjourney (an AI art tool), even though the storyline was human-written. In the landmark decision of Naruto v. Slater, 888 F.3d 418 (9th Cir. 2018), the court famously held that a monkey could not own copyright in a photograph it had taken, underscoring the requirement of human authorship. More recently, in 2023, the Copyright Office reiterated that copyright protection would not extend to AI-generated content lacking substantial human input. This strict interpretation may serve well for now, but as AI becomes more autonomous, it could leave a large swath of creative output unprotected, unless Congress steps in.
Europe, particularly the European Union, has been more proactive in creating a legal framework for AI. While there is still not a harmonised IP rule across member states, the European Union proposed Artificial Intelligence Act is a major step forward in the direction of AI legislations. Though not an IP law per se, the Act sets out principles of accountability, transparency, and traceability in the use of AI. These principles could form the backbone of future IP regulation. Moreover, countries like Germany and France have also initiated policy-level dialogues on how AI can fit into their IP systems. The European Parliament’s 2020 Resolution on Intellectual Property Rights for the development of AI urged the Commission to consider new categories of rights for AI-assisted works, though nothing has been codified yet.
Australia and Canada have taken cautious but interesting steps. In 2022, Australia’s Federal Court initially ruled in Thaler v. Commissioner of Patents [2021] FCA 879 that an AI could be an inventor under the Patents Act 1990. But this was overturned by the Full Federal Court in 2022, reaffirming the requirement of human inventorship. Canada’s IP Office recently updated its patent examination guidelines to require “a natural person who made a material contribution” in the conception of the invention—effectively excluding standalone AI inventorship.
These comparative examples show a spectrum of approaches: the UK is pragmatic, the US is conservative, the EU is regulatory, China is opportunistic, and Australia is cautious. India, in contrast, is passive. But passivity may be the most dangerous stance of all, because it leaves creators, developers, and investors in a state of uncertainty. Worse, it may allow powerful entities to exploit loopholes in the absence of clear rules.
So what can India learn? First, that legal clarity is not the enemy of innovation—it is its enabler. Second, that adopting a one-size-fits-all approach will not work. And third, that a patchwork of judicial decisions cannot substitute for comprehensive legislative reform.
CONCLUSION: REWRITING THE IP SCRIPT IN THE AGE OF ALGORITHMS
As we reach the end of this deep dive into the intersection of artificial intelligence and intellectual property law, one thing becomes abundantly clear this is not a moment of mild disruption. It is a full-blown reordering of how we conceive, create, and credit intellectual works. Copyright and patent law have historically been anchored in human creativity and inventiveness, now stand at a cliff’s edge. The law student’s instinct is to reach for precedent, parse the text, and find a bridge between principle and practice. But here, the bridge hasn’t been built yet—and maybe that’s the point. Maybe we’re living through the construction phase of a legal edifice for a new technological reality.
Artificial intelligence is not just a tool, it’s becoming a collaborator. It doesn’t merely extend human creativity it can, in certain contexts, replace it or even exceed it. From generating classical symphonies and bestselling novels to discovering new drug molecules and engineering designs, AI’s fingerprints are all over what we once thought to be the exclusive domain of human imagination. And yet, our laws are still built around an anthropocentric model. We still think of an author as a person hunched over a desk and of an inventor as a scientist in a lab. AI has exposed the limitations of these images.
The problem is that this legal lag isn’t just academic. It has real consequences. It creates uncertainty in the marketplace. It disadvantages Indian innovators in the global IP ecosystem. It risks overprotection in some areas and underprotection in others. And it opens the door to exploitation where either the rights of human creators are swallowed up by powerful platforms or AI-generated works are pirated freely due to the absence of legal clarity.
The courts, for their part, have tried to be pragmatic. In copyright, they’ve slowly shifted from rewarding labour to rewarding originality, a shift that began in Eastern Book Company and has been refined in later decisions. In patent law, they’ve underscored the need for ingenuity and public interest, as seen in Bishwanath Prasad Radhey Shyam. But there’s a ceiling to judicial reform. Statutory silence can only be filled with interpretive creativity for so long. Eventually, Parliament must step in. It must reimagine the Copyright Act and the Patents Act for an age where the “author” might be a neural network and the “inventor” a self-learning algorithm.
At the same time, legislation must not be reactionary. We must resist the temptation to overregulate out of fear. The goal is not to create a new legal person out of code, but to assign accountability, define ownership, and preserve incentive structures. Perhaps the way forward lies in attributing rights to human agents who control the AI. Perhaps we need new, sui generis rights for AI-generated outputs. Or perhaps we must rethink our entire IP framework to focus more on user rights and public benefit than on rigid categories of authorship and inventorship.
What we certainly need is dialogue—between lawmakers, technologists, judges, academics, and most importantly, creators. Artists and inventors are not mere spectators in this debate; they are the stakeholders with the most to lose or gain. If AI becomes the new brush or microscope, we must decide who wields it, who owns the output, and who gets to profit from it.
The problem of AI and IP is not a glitch. It’s a generational challenge, a redefinition of legal philosophy, and an opportunity to future-proof our legal system. If we don’t act soon, we may find ourselves stuck in a copyright regime that protects no one, and a patent system that rewards no real inventors. But if we get it right, we can write a new chapter in intellectual property—one where law and technology evolve together, not in conflict, but in collaboration.