Time to talk about ownership of AI-generated intellectual property assets

Intellectual property (IP) systems have been in place since the Middle Ages and encourage skilled, innovative technicians by granting monopolies within certain industries. For hundreds of years, intellectual property policy has been dictated by the imperative to reward human creativity or ingenuity that brings new creative works and useful inventions to society in return for disclosing those works and inventions in order to advance progress promote – the so-called “bargains.”

The advent of machine learning and modern artificial intelligence (AI) is now calling this paradigm into question. Computers have become more powerful, and as they acquire higher-order brain functions through machine learning, they have developed the ability to do activities that humans would otherwise consider original or inventive, and create copyrighted works and patentable inventions. Quantum computers are guaranteed to accelerate this trend.

In 2021 we saw the first global court rulings addressing the question of whether a non-human can be an inventor within the meaning of patent law.

But who do these in silico creations belong to? The answer is far from clear, in part because of the competing policy goals underlying IP systems. While intellectual property protection is designed to reward people for their creativity and ingenuity in promoting such behavior, intellectual property protection is also designed to drive technological advancement, regardless of how it comes about. IP statutes are made by people for people – but does that have to be the case?

The past year has brought this issue to the fore, both in patent and copyright law.

Patents: can AI be an inventor?

In 2021 we saw the first global court rulings addressing the question of whether a nonhuman can be an inventor within the meaning of patent law. A patent application has been filed in 17 countries with an AI inventor named DABUS (Device for Autonomous Bootstrapping of Unified Sentience). DABUS was founded by Dr. Stephen Thaler and is named as the owner of the patent applications. Patent offices had to check whether a patent could be granted with DABUS as the named inventor. So far, four countries have participated with mixed results.

In July 2021, DABUS scored two victories. Initially, the South African Patent Office issued a patent listing DABUS as the inventor, although no reason was given, as South Africa grants patents without substantive examination (and this patent can still be challenged in court).

Shortly thereafter, the Australian Federal Court of Justice came to the same conclusion under Australian law, stating that patent law was not a requirement for an inventor to be human. Motivated by the need to encourage and reward technological innovation, the Australian court found that the term “inventor” was undefined and its common meaning (like other agent names such as “computer” or “dishwasher”) did not exclude nonhumans. After concluding that DABUS was the inventor, the court found that Dr. Thaler was the owner of the invention as he derived the title from the inventor DABUS. In order to arrive at that conclusion, the court argued that it was not necessary for an inventor to be a legal entity capable of assigning rights in order to conclude that the proprietor’s title “derived from the inventor under Australian Patent Acts ” would. It will be interesting to see whether this rationale, which justifies the transfer of rights from a non-human inventor to a human successor in title, is applied in other jurisdictions.

Courts in the United Kingdom and the United States have reached a different conclusion on inventiveness. The UK Court of Appeals concluded that the character and duties of an inventor require that he be human. Machines have no legal personality and cannot have rights, nor can they transfer rights to their owners. A machine cannot make a statement that it is the true inventor of an invention. In the United States, the US District Court for the Eastern District of Virginia referred to the definition of an inventor in US patent law as a “natural person” who must be a “natural person” under US law and to the human-centered requirement an inventor to indicate his “belief” in his inventorship.

No Canadian agency has commented on this matter to date, and while Canadian patent law is most closely modeled on US and UK law, it is clear that the prospects in these jurisdictions will carry weight in any future case. However, as in Australia, there is no definition of “inventor” in Canadian patent law.

Copyright: Can AI be an author?

AI also raises new questions about copyright, including the authorship and ownership of AI-generated works. Under current copyright law in Canada, it is unclear whether AI-generated works are protected by copyright. Copyright protects works that are the result of an author’s exercise of skill and judgment. The default rule is that the author is the first owner of the copyright (subject to certain exceptions). There is no definition of “author” in Canadian copyright law, but copyright law suggests that an author must be an individual.

Today, AI systems are able to create works that – partially or even completely – arise independently of human intervention. This development calls into question the established legal doctrine, which has understood and defined authorship as an act of expression emanating from a person. Whether or not AI-generated works are copyrighted, and if so, who owns the rights to the work, is an important issue affecting public order and the Canadian economy.

In July 2021, the Canadian government published a consultation paper calling, among other things, for a modern copyright framework for AI in Canada. Regarding the authorship and ownership of AI-generated works, the consultation paper suggested three possible approaches:

• The first approach is to make fully AI-generated works non-copyrighted. This approach reflects the legal situation in a number of countries, including Australia, where copyright (as opposed to patented inventions) only protects works created by a human author, not machine-generated works.
• The second approach is to ascribe the authorship to the person or person who caused the work to be created (but not to the AI ​​who actually created the work). This approach to AI-generated works has been implemented through legislative changes in several common law jurisdictions, including the United Kingdom, Ireland and New Zealand.
• The third approach is to allow copyright protection of AI-generated works, but consider them to be “authorless”. Under this approach, no moral rights would likely be attached to AI-generated works, meaning that no person would have the right to have their name attributed to the work as the author or to maintain the integrity of the work.

While adopting any of the approaches above would help clarify the legal rights to AI-generated works under Canadian law, each one has a significant impact on the economy and public order. This is especially true given the importance of AI in the modern economy and the public interest in promoting the development and use of AI in Canada. It remains to be seen what approach Canada will take.

Impact of increasing recognition of AI-generated intellectual property

Important consequences result from the decision to formally recognize and reward AI-generated creations in intellectual property systems.

If AI-powered service providers cannot protect their creations, they lack the bargaining power in commercial agreements. In this situation, commercial affairs must be structured in such a way that human contributions are made as the basis for asserting or obtaining copyright or patent protection. If patent and copyright systems are seen as inadequate, companies can also choose to keep their innovations as trade secrets instead of making their intellectual property public. Legal differences will make these business evaluations difficult.

Conversely, when AI-generated creations are patented or copyrighted, the owners of the best performing AI are empowered to take control of entire areas of intellectual property, potentially sparking an intellectual property arms race in which humans compete against machines. At this early stage, when AI is only occasionally inventive, it is difficult to see where that empowerment could lead. Canadian IP policy decisions on AI-generated creations are likely driven by a desire to attract, not discourage, investment in AI research and development in Canada. These decisions require thought and creativity and (dare we say) a human touch.

https://www.lexology.com/library/detail.aspx?g=3664f742-a72a-4661-9e97-82637eb8a993