Humans, You are Responsible for Your Machines

By
Ava Chisling
-
May 31, 2017

Since the painting above was created by a modified welding machine named eDavid, we wondered about the legalities of other non-human works, whether it’s art by eDavid, fake news on Facebook, or the monkey selfie. Who owns the rights to these works — and more importantly, who is responsible for them? We asked a leading expert in AI and robots, advanced technology and IP law to clear things up.

If you’re going to have an informed discussion about AI, robots, creativity and the law, Dr. Yanisky-Ravid is the perfect person to call. She is a law professor and a research fellow at Yale, a visiting professor at Fordham Law and a senior faculty member at Ono Law School in Kiryat Ono, Israel. Dr. Yanisky-Ravid has lectured at Harvard Law, Columbia University, Lausanne University and works with WIPO (the World Intellectual Property Organization), the Swiss Institute of Comparative Law and the Oxford Center for Competition Law and Policy.

When we say something is an “AI system” what does that mean?

Artificial intelligence systems are creative, unpredictable, independent, autonomous, rational, evolving, capable of data collection, communicative, efficient, accurate, and have free choice among alternatives. Similar to humans, AI systems can autonomously create and generate creative works. The AI industry has not only rapidly become an inevitable part of our present, but it is expected to further develop with the industry estimated to grow to $70 billion by 2020.

So, AI systems are artists? If yes, is their work copyright-protected?

Surprisingly, AI-like systems are already producing original works in almost every copyrightable medium, such as music, poetry, literature, and news. Indeed, today, it’s almost impossible to imagine any kind of art developed without using at least some digital means. Eventually, as we have argued in our works, we are confident that automated systems will replace both creators and producers of numerous types of works, products, and services.

As we face the 3A era of Advanced, Automated, and Autonomous AI systems, and as AI systems begin to create art, copyright laws become outdated and thus irrelevant. I think, as a result, the concept of originality and art as necessarily human-based is very likely to lose its meaning. We should adopt an objective tool, a test to make sure the works do not infringe on another entity’s rights.

You disagree with the now infamous “monkey-selfie ruling.” Where do you stand on animal, robot and other non-human creations and the law?

The idea that a monkey could own art and that robots could go to jail are, on the one hand, very logical if we understand the human–like nature of AI systems. But, on the other hand, they are not practical because we can’t rely on their accountability and the somewhat ridiculous and incompatible constitutional justifications underlying intellectual property (to motivate creators and inventors). Therefore, I think copyright laws are unprepared to deal with copyright issues and they should be replaced in these cases with other tools. I suggest a different solution.

In a recent article I wrote, I propose the adoption of a new model for the accountability for works generated by AI systems: the AI Work Made for Hire (WMFH) model, which views the autonomous AI system as a creative employee or independent contractor of the user. Under this proposed model, ownership, control, and responsibility would be imposed on the humans or legal entities that use AI systems. This model accurately reflects the realities and human-like features of AI systems. It is justified by the theories behind copyright protection, and it serves as a practical solution for the fears behind AI systems. In addition, this model unveils the powers behind the operation of AI systems and, hence, it efficiently imposes accountability on clearly identified persons or legal entities.

Tell me more. If jail is out for the robot, what happens if an AI-created work breaks the law? Are we going to sue or jail the programmer? The coder? The owner?

The question of accountability is very complicated when it comes to AI systems because of the multiple contributors. In the article I mentioned earlier, we address the question of who should enjoy the benefits of copyright protection, as well as who should be responsible for the infringement of rights and damages caused by AI systems that independently produce creative works. We argue against the imposition of these rights and responsibilities on the AI systems themselves or on the different stakeholders (mainly the programmers who develop such systems).

Because of the autonomous, evolving, and creative nature of the AI system, programmers often cannot predict the outcome. In the same way the manufacturer of a camera, computer, or a piano does not own the picture, story, or rhythm created with these tools, the programmers should not be considered the owner of works generated by creative AI systems. Like Stephen Hawking said: “The short-term impact of AI depends on who controls it; the long-term impact depends on whether it can be controlled at all.”

So humans are responsible for non-human works?

We concluded that humans are the employers and thus the owners of artworks created by AI systems (their creative employees). This new AI Work Made For Hire model we propose entitles users (i.e. humans) ownership to the creative artworks, along with responsibilities for the AI systems.

How about this one? eDavid made this artwork, too. Does it own the rights to it? What if it infringed upon your work? You can see eDavid in action here.
AI Created Work
/ e-david.orgProf. Deussen, Thomas Lindemeier, Work group Visual Computing, University of Konstanz

If we were having this conversation in 10 or 20 years, what will the world look like from an AI and IP point of view?

I can see it clearly. All works now considered as art and regarded as inventions and patents will be created by automated AI systems using 3D printers and other tools. No copyright and patent laws will exist. No patent applications. Licenses to use the products will be done by electronic contracts and the payment will be in bitcoins. Sanctions will be done electronically. Avoiding infringements and counter rights will be done electronically. Dispute resolution will be done by digital alternative dispute resolution AI systems, which means courts will close. The only question remaining is: where are wein these future scenes?

Dr. Yanisky-Ravid has published numerous articles and has won numerous awards and scholarships for her work. Recently, her article, Generating Rembrandt: Artificial Intelligence [AI], Accountability and Copyright — Our New Sophisticated Workers are Here — A New Model (co-authored with Samuel Moorhead), was nominated as the visionary article of 2017 in intellectual property. Her article on the right to privacy and the Balloon Theory was deemed to be one of the best law review articles related to entertainment, publishing and/or the arts published in the US, in 2014. Dr. Yanisky-Ravid has also published a book, Intellectual Property at Workplaces: Theoretical and Comparative Perspective, which suggests a new way of thinking about IP in the workplace.

Ava Chisling

Ava is an award-winning lawyer and editor who counsels creative types, writes about pop culture/tech+law and sometimes creates ad campaigns. She is Quebec counsel for Momentum Law.