Artificial intelligence (AI) and intellectual property (IP), a call for action

by: in Law
Artificial intelligence blog_Law blogs Maastricht

“Whereas from Mary Shelley’s Frankenstein’s Monster to the classical myth of Pygmalion, through the story of Prague’s Golem to the robot of Karel Čapek, who coined the word, people have fantasised about the possibility of building intelligent machines, more often than not androids with human features”.

Seldom have we seen a document originating from a body of the European Union being introduced in the literary-romantic way as the Motion for a European Parliament Resolution A8-0005/2017 (of January 27th, 2017) by the Committee on Legal Affairs of the European Parliament.

This report on Civil Law Rules on Robotics constitutes an acknowledgement of the likely upcoming event that “ever more sophisticated robots, bots, androids and other manifestations of Artificial Intelligence (“AI”) unleash a new industrial revolution”, and that such an event will affect all spectra of society.

Overly optimistic view
Due to its inherent link to society, this development will also affect all sorts of law extensively. In the report it is stated that even though there are no legal provisions that specifically apply to robotics, “existing legal regimes and doctrines can be readily applied to robotics, although some aspects appear to call for specific considerations”, which in our view might be overly optimistic as to the capabilities of current legislation. But at last, there has been a call to legislate these technologies. Preferably, as the report states, without restraining innovation and whilst considering the ethical and legal implications and consequences.

In the report the Committee calls on the Commission to come up with a “horizontal and technological neutral approach to intellectual property” with regard to artificial intelligence, or a “balanced approach to IPR’s when applied to hardware and software standards and codes that protect innovation and at the same time foster innovation”. The meaning of this formulation is in our view quite ambiguous. From an intellectual property perspective, to us the most interesting aspect of the report however, is that the resolution asks for the elaboration of criteria for “own intellectual” creation for copyrightable works produced by computers or robots.

The first question that came to our mind was why they only specifically demand elaboration on criteria for this, and not for example in other fields of intellectual property such as Patent Law. What about the other fields of IP law? And why should we or should we not want Artificial Intelligence generated copyrighted works or patents? In this blog we will limit ourselves to observe and indicate some potential issues that may arise.

Rethink the role of copyright law
As human creators are increasingly assisted by AI systems or co-creating works with autonomous machines, whereas also the creative input can potentially also come from computer programs, we may indeed have to rethink the role of copyright law. In some cases the creative process is even left entirely up to such systems. Some issues that arise challenge the very foundations underlying IP laws. Questions such as how to provide incentives and protection of investment for artists and the industries that work with such systems, but also what kind of ownership results from works (partially) created by AI machines. One of the issues with ownership is that these are not even new sources of creative works in some cases, but they depend highly on access to works created by others and require massive amounts of input-data (which can be subject to varying IP regimes). This input must remain at an economically affordable cost if we want to incentivise the AI revolution and ensure the interoperability of network-connected autonomous robots. But also the other way around needs important considerations, as the current legal framework is insufficient on “data protection and ownership due to the (expected massive) flow of data arising from the use of robotics and AI”.

It leaves us to wonder what should happen to traditional IP law concepts as ‘inventiveness’, ‘original’, and ‘creator’ in this new evolving environment of machine creation. For example, currently, only the UK has “an explicit regulation for (some) computer generated works, though the strict equivalence with human works may give overly generous protection to works that can increasingly be produced with almost no effort and in quantities previously unimaginable.”

Complexity in ownership
In patent law one of the crucial issues could be the definition of the ‘inventor’, do we want this to include a machine? As Stephen Hawking put it: “the short-term impact of AI depends on who controls it; the long-term impact depends on whether it can be controlled at all.” Legally speaking, it can lead to a complexity in ownership, especially if multiple contributors are involved in the development of an AI system. In infringement cases we have another problem as there currently is no legal framework that provides for a machine or program as an infringing party, meaning robots cannot be held liable for acts or omission that cause damage to third parties. Even though the human creator/operator of the robot can be held liable, this leaves the issues to identify who the owner is and the scenario where the robot takes autonomous decisions. Some experts[8] have argued that “computers should be allowed to patent their inventions”. They claim that otherwise there will be less innovation and individuals will claim credit for inventions that are not genuinely theirs (but the creation of an AI system). Another question, apart from the question of patentability and ownership, is whether or not AI inventions should be considered prior art. If AI’s start spitting out random inventions which are not patentable for the mere reason they are AI generated, in theory eventually every future human made invention could become unpatentable as the invention has been covered by AI generated prior art, potentially rendering the patent system useless in the future.

To solve some of the issues the Committee calls for ‘common Union definitions’ and a ‘comprehensive Union system for registration’ with criteria for classification of robots.Legal policy should “encourage helpful innovation, generate and transfer expertise, and foster broad corporate and civic responsibility for addressing critical societal issues raised by these technologies.” The challenge will be on the EU legislator to develop a comprehensive legal framework in this rapidly evolving world of AI systems. Or will we learn that can they cannot be controlled at all?!

 Written by Anne de Heijde & Raynor van Eijck

  Published on Law Blogs Maastricht