How AI Can Cure the Big Idea Famine
What Good is the Law?
"It was just kind of a crisis of humanity. People just weren't really stepping up to the magnitude of the problem. They weren’t really human for a moment."
- James Love, Fire in the Blood (2013)
In 2001, MIT Technology Review published an article provocatively titled “In Africa, Patents Kill.” 1 The article described the HIV/ AIDS crisis that had engulfed Africa: over 25 million patients, 2.4 million annual deaths, life expectancy reduced to 27 years, and little by way of medical help. This was a time when AIDS was a death sentence, an ever-present Damocles' sword that did not discriminate between rich and poor, young and old. Treatment was hard to come by because the disease was taboo and fears that it could spread to the doctors and nurses treating patients.
Help did not reach those who needed it most; it had to be purchased at a premium in the form of colorful capsules. The purveyors of these capsules went by many names—Pfizer, GlaxoSmithKline, Merck, Novartis—but the one thing they all had in common was the right to manufacture and distribute the various antiretroviral drugs needed for treating AIDS.
Monopoly bred exploitation. Not only did these companies sell medicines at prices that made them unaffordable to most patients in Africa, they threatened legal action against those attempting to sell generic versions at lower prices. In Ghana, Glaxo blocked the importation of life-saving generics by writing to the Indian generic-manufacturer, arguing that the importation violated Glaxo's patent rights and was, therefore, illegal. As the journalist Donald McNeil notes, the attitude of the Big Pharma companies was, “You fight our patent monopolies, we will make sure you die.” 2
And here’s the truly damning part: The law supported such exploitation. Companies like Glaxo were correct in arguing that importation of generic versions of their drugs infringed on their patent rights. Patent laws were passed to reward inventors who develop innovative technologies that solve the world’s problems. “To promote the progress of science…” reads the United States Constitution in the context of patent law. The rationale goes that without exclusive rights, inventors have no incentive to conduct R&D and invest time and money to advancing science and technology through their inventions.
The millions of Africans who died from HIV and AIDS were just martyrs for the greater good of encouraging scientific innovation and progress in society.
Halfway across the world, in 2013, Aaron Swartz took his own life. Swartz had worn many hats: computer programmer who co-designed RSS and the Creative Commons, political activist instrumental in fighting the Stop Online Piracy Act (SOPA) and the Protect Intellectual Property Act (PIPA), and entrepreneur who built Reddit but wasn’t in it for the money. He also happened to be a staunch supporter of the open access movement.
Swartz committed suicide after being targeted and prosecuted by the FBI for illegally downloading academic journals and books from the online repository, JSTOR. The government sought millions of dollars in fines and a maximum jail sentence for Swartz; but the zeal with which he was prosecuted made no sense. His actions had not caused lasting, irreversible harm to any person or organization; after all, he had just downloaded academic content from the Internet.
It was Swartz’s ideology that was being persecuted, not his actions or “crimes.” The 2000s had been tumultuous for the copyright industry, a period that saw the rise of file-sharing websites such as Napster and The Pirate Bay along with the very public failure of SOPA and PIPA. These incidents were evidence of the growing public sentiment against copyright monopolies. Stuck in a relentless game of whack-a-mole, the industry felt threatened, even vulnerable. It had spent millions of dollars in lawsuits, and despite winning several cases, online piracy refused to go away.
Swartz was the star of this team of pirates. He was not just committing silent acts of piracy for personal benefit, he was bringing the challenge to the system, the same system that had enabled the copyright industry to make mammoth profits by pricing its content at will. Why should the public pay for reading and understanding the law? Why should taxpayers be compelled to pay exorbitant amounts for accessing publicly-funded research? Why should knowledge be accessible to only the elite few who could afford it? Swartz asked the questions; the industry had no answers.
So, they decided to stop Swartz and make an example of him. Armed with copyright laws and archaic Internet laws, the government went after him with all its might. Swartz hung himself the day before his trial. The law had claimed another martyr.
The two stories above have many differences. They come from contrasting parts of the world: the so-called Third World and the United States. The issues seem unrelated—the former about access to life-saving medicines, and the latter about books, academic papers, and other creative works. The first story is about the survival of an entire continent, the second is about a man's personal battles against an unjust system.
Yet the stories have many common threads and offer a bunch of lessons to mull over. First, the law, with all its historical baggage and lack of foresight, can be unjust. Second, unjust laws need to be challenged because otherwise they can lead to victimization and alienation of certain sections of society. Third, challenging unjust laws can cause the system to resist and even strike back because people, especially those in power, are averse to change. And, finally, technology can be the catalyst for change whether by allowing reverse-engineering of patented drugs for manufacture of generic medicines or by enabling file-sharing across continents.
This essay is about how technology—specifically, AI—can give unjust laws, monopolies, and the gatekeepers who profit from them a run for their money.
The Philosophy of Open Access
A famous anecdote about Charles Dickens’ fight for copyright protection for his books goes something like this: In the mid-1800s, American copyright law did not protect works created by foreigners. As a result, Dickens and other non-American authors found their works getting copied and sold at lower prices in the United States. To Dickens’ chagrin, even though his works became increasingly popular, he lost royalty since cheap, pirated copies were replacing the market demand for the more expensive originals.
In 1842, Dickens traveled to the United States where he delivered a speech about the need for an international copyright law and even tried lobbying the Congress on this issue. Public sentiment at the time was deeply rooted in values such as the freedom of press, and Americans saw no reason to pay for such content. Dickens returned to Britain with the label of a “mere mercenary scoundrel.” It wasn’t until 1891, five decades later, that American law began protecting the works of foreign authors under the International Copyright Act.
This anecdote highlights an important point: Intellectual property laws generally reflect the wants of society. Through the 1800s, American society benefited from not providing copyright protection for works of non-citizens because this allowed American citizens and the press to access and use foreign works cheaply. By 1891, as foreign trade grew, the American government realized the problems with this approach, with the most significant being that foreign countries, such as Britain, also did not protect works of American authors. That, coupled with international pressure, led to the reform that was the International Copyright Act of 1891.
Intellectual property laws have come a long way since then. We live in maximalist regimes today where intellectual property laws govern almost every facet of our lives. Our books, movies, smartphones, cars, shoes, medicines, and even food in some cases, are protected by at least one branch of intellectual property law. The reason for this expansion is that without intellectual property protection, authors, inventors and creators of any kind would have no incentive to make something new because their creations would just be copied by others. This is known as the incentive theory, which essentially states that intellectual property laws incentivize and reward creators for their contributions to society. Without exclusive rights, creators would have no recourse against those pirating and distributing their creations, which would prevent them from recouping their investment of time, effort, and money. This would lead to a net loss of innovation and creativity in society.
The incentive theory has come under siege due to proliferation of technologies such as the Internet and AI that have changed the scientific and creative sectors in a plethora of ways. Economies of scale and manufacturing tools such as 3D printing have reduced production costs, research has become easier due to constant availability of information, marketing and distribution costs have been minimized by the Internet, big data has allowed creators to understand and target consumers better, and more avenues for revenue generation have been developed. All these factors have given birth to the idea that the incentive theory, as traditionally understood, no longer bears merit.
Supporters of open access are not against the incentive to create; neither are they against the concept of intellectual property laws themselves. Instead, they argue that technological progress has generated opportunities for creators to benefit from their creations without having to price their products and services excessively. The equation between access and profit no longer has to be a zero-sum game.
Accessibility has become a fundamental need and people are increasingly growing tired of exploitation of monopolies by gatekeepers.3 As evidence of how deeply entrenched this problem has become: Harvard University, which receives some of the largest endowments in the world, released a statement in 2012 encouraging its faculty members to publish with open access journals because the university could not afford subscriptions from journal publishers.4 Likewise, in India, the Supreme Court5 was compelled to intervene in a case where Bayer was selling its patented anti-cancer drug at over INR 2,80,000 (~ USD 4,100) a month, which made the drug available to only about 2 percent of the patient population. Peer-to-peer file sharing, a burgeoning generic pharmaceutical industry, and the rise of platforms such as LibGen and Sci-Hub are all indicators that our current approach toward intellectual property is no longer consistent with public perception.
At no point of time in the history of intellectual property laws has the mismatch between government attitude and public demand appeared to be so stark. While people want to migrate to a model of openness and improved access, governments are adopting extreme measures such as criminalizing infringements, introducing newer forms of intellectual property such as data exclusivity, and arm-twisting countries through international treaties and sanctions into complying with higher standards of protection. The philosophy of open access is rooted against the exploitation that has become the natural consequence of entities being conferred with exclusive rights for extended periods of time. Unbridled intellectual property-based monopolies have fragmented society into haves and have-nots when it comes to accessing knowledge, culture, and lifesaving technologies.
These problems have also been noted by Joi Ito in his manifesto, Resisting Reduction, wherein he writes that the primary currencies of our civilization are money and power, and the goal often is to accumulate both at the expense of society. Intellectual property law has proven to be one of the greatest enablers of such accumulation. Similarly, in Big Idea Famine, Nicholas Negroponte singles out Apple for having an egregious track record of not making its research available to the public yet benefiting enormously from this research. “You cannot keep skimming the cream off the top, without doing some basic, open research that is widely shared. Open and shared are the key words,’’ he wrote.
The open access movement might just find the unlikeliest of allies in AI.
After decades of research, experiments, breakthroughs (hello, Deep Blue), uplifted hopes and dashed spirits, one could finally say that the AI revolution is here. In autonomous cars and world champion-beating programs, legal research and chatbots, surgery and autonomous weapons systems, AI is steadily pervading every sphere of society.
The creative and scientific domains will not escape unscathed. The next revolution in scientific innovation and development of creative works is likely to come in the form of AI that can autonomously invent and produce creative content. Early signs of this are already visible: From Ray Kurzweil's Cybernetic Poetic6 to Stephen Thaler’s Creativity Machine,7 AI has demonstrated the capacity to “learn” from existing material and produce new, original output that can stand as an independent work. The process is analogous to how humans learn and create new works by extrapolating from their learnings.
It is not difficult to imagine a scenario in the near future where most inventions and creative works will be developed by AI. The big question is, who should own the intellectual property rights to such creations?
The most intuitive answer is that the developer of the AI should own all the works that are generated by the AI. This response is based on the rationale that the initial effort of coding, training, and improving the AI was invested by the developer, who should be entitled to the creative outputs of the AI. If it had not been for the developer, the work would have never come into existence. Further, if the developer is given ownership of the intellectual property that is created by the AI, it would incentivize her and other developers to build better AI systems.
This approach is being implemented in countries such as the United Kingdom,8 New Zealand,9 and Ireland.10 The copyright legislations of these countries have been amended to state that copyright in computer-generated work would vest with the person who made the necessary arrangements for such works to be created. In most cases, such a person would be the developer, since, by developing the AI, she made the “necessary arrangements” for a copyrightable work to be created. This position is an extension of the work-for-hire doctrine, whereby the AI system is treated as an employee, while the developer is considered the employer who owns the intellectual property rights to all works created by the employee in the course of her employment.
There is no denying that this argument has some merit and is also consistent with the incentive theory. However, this approach fails to recognize that the actual “effort” of developing the work, whether it be a scientific invention or some creative content, does not arise from the developer of the AI. The role of the developer is limited to laying down the groundwork on which the AI can generate the creation. Beyond this point, there is no originality or novelty, which are requisites for copyright-ability and patentability, respectively, exercised by the developer in the work or invention itself. The process is similar to a professor teaching the fundamentals of biochemistry to a student which the student builds on to develop a new pharmaceutical product. One would be hard-pressed to say that the professor should also be allowed to claim rights in the student’s invention.
Concerns about the incentive theory are not particularly pressing, either. Even if developers are not given rights to inventions and creative works produced by their AI, the market value of an AI with such capabilities would be adequate incentive for developers to continue making such systems. It is for this reason that developers of tools available in the market today, such as 3D printers and photo editing software, do not feel the need to lay claim to works that are created using such tools. Their incentive comes from the value that the tools themselves generate.
The second option would be vesting the intellectual property rights to the AI’s creations with the owner/user of the AI system. Simply put, if I ask an AI system on my computer to write an original poem about astronauts or invent a new type of shoe-sole that optimizes surface friction, and the AI system does this successfully, then I should own the intellectual property rights to these creations because I posed the AI system with the problem and set the parameters within which the system worked. It was my ingenuity in framing the right kind of question and in rejecting false outputs that led to the eventual creation. Surely, my efforts and intellectual labour deserve to be rewarded.
Like the first proposition, this solution has a prima facie appeal to it. It can be argued that innovation in society would be maximized if people are given powerful AI-driven devices along with the guarantee that the labor they spend in developing new creations using such devices would be rewarded with them getting exclusive rights to use and commercially exploit these creations. It would also encourage developers to build better AI systems because users would then be willing to pay higher prices for such systems knowing that any intellectual property developed by the system would belong to them.
However, this explanation misattributes the source of the creativity or ingenuity, which is not the user but the AI system itself. Going back to the professor-student analogy, the professor cannot claim that she developed a new invention or song if she merely set the parameters or instructions within which the student had to operate. It is for this reason that current intellectual property laws afford no protection for mere identification of a problem or idea. It is the actual expression (in case of copyright) and the inventive solution (in case of patents) that is protected.
Before moving to the most convincing solution, there is a third position that is worth mentioning. There is a school of thought which believes that the AI system itself should be accorded with intellectual property rights to its inventions and creative works. Some have even drawn parallels with the case of Naruto v. David Slater,11 popularly known as the monkey-selfie case, wherein it was argued that the copyright to a selfie taken by a macaque should vest with the macaque. Through several stages of appeal, courts consistently held that under American copyright law, only humans could be authors of creative works. Non-human entities, such as macaques and even computers, cannot be “authors” under copyright law.
While this theory does correctly recognize the origins of the creativity/novelty, it is difficult to foresee a scenario where the AI system will meaningfully be able to exercise and defend its intellectual property rights. Most experts would agree that at least for the next few decades, AI systems would remain narrow and constrained to perform a limited set of tasks. This solution, thus, does not address these practicalities.
Which brings us to the most befitting answer: No one owns the intellectual property rights to AI-generated creations, and these creations fall into the public domain. This may seem unpalatable at first, especially since intellectual property laws have played such a fundamental role in our society so far. We have been conditioned to a point where it seems almost unimaginable that some creations should directly enter the public domain upon their birth.
But, doctrinally, this is the only proposition that stays consistent to extant intellectual property laws. Works created by AI have no rightful owner because the application of mind to generate the creation, along with the actual generation of the creation, would entirely be done by the AI system. Human involvement is ancillary and is limited to creating an environment within which such a creation can take form.
This can be better understood through a hypothetical example: If an AI system were to invent a groundbreaking pharmaceutical ingredient which completely treats balding, then the system would likely begin by understanding the problem and state of prior art. It would undertake research on causes of balding, existing cures, problems with existing cures, and whether its proposed cure would have any harmful side effects. It would also possibly combine research and knowledge across various domains, which could range from Ayurveda to modern-day biochemistry, before developing its invention.
The developer can lay as much stake to this invention as the team behind AlphaGo for beating Lee Sedol at Go. The user is even further detached from the exercise of ingenuity: She would be the person who first thought, “We should build a Go playing AI system,” and direct the AI system to learn Go by watching certain videos and playing against itself. Despite the intervention of all these entities, the fact remains that the victory only belongs to AlphaGo itself.
Doctrinal issues aside, this solution ties in with what people need from intellectual property laws: more openness and accessibility. The demands for improved access to medicines and knowledge, fights against cultural monopolies, and brazen violations of unjust intellectual property laws are all symptomatic of the growing public discontent against strong intellectual property laws. Through AI, we can design legal systems which address these concerns and reform the heavy handed approach that has been adopted toward intellectual property rights so far.
Tying the Threads Together
For the above to materialize, governments and legislators need to accept that our present intellectual property system is broken and inconsistent with what people want. Too many people are being deprived of basic amenities such as medicines, patent trolls and patent thickets are slowing innovation, educational material is still outside the reach of most people, and culture is not spreading as widely as it should. AI can provide an opportunity for us to redefine this paradigm—it can lead to a society that draws and benefits from an enriched public domain.
However, this approach does come with built-in cynicism because it contemplates an almost complete overhaul of the system. One could argue that if open access for AI-generated creations does become the norm, then innovation and creativity would suffer as people would no longer have the incentive to create. People may even refuse to use their AI systems, and instead stick to producing inventions and creative works by themselves. This would be detrimental to scientific and cultural progress and would also slow adoption of AI systems in society.
Yet, judging by the pace at which these systems have progressed so far and what they can currently do, it is easy to imagine a reality where humans developing inventions and producing creative works almost becomes an afterthought. If a machine can access all the world’s publicly available knowledge and information to develop an invention, or study a user’s likes and dislikes while producing a new musical composition, it is easy to see how humans would, eventually, be pushed out of the loop. AI-generated creations are, thus, inevitable.
The incentive theory will have to be reimagined, too. Constant innovation coupled with market forces will change the system from “incentive-to-create” to “incentive-to-create-well.” While every book, movie, song, and invention is treated at par under the law, only the best inventions and creative works will thrive under the new model. If a particular developer’s AI system can write incredible dialogue for a comedy film or invent the most efficient car engines, the market would want more of these AI systems. Thus incentive will not be eliminated, it will just take a different form.
It is true that writing about such grand schemes is significantly tougher than practically implementing them. But, for any idea to succeed, it must start with a discussion such as this one. Admittedly, we are still a moonshot away from any country granting formal recognition to open access as the basis of its intellectual property laws. And even if a country were to do this, it faces a plethora of hoops to jump through, such as conducting feasibility-testing and dealing with international and internal pressure. Despite these issues, facilitating better access through AI systems remains an objective worth achieving for any society that takes pride in being democratic and equal.