“Everything that can be invented has been invented.” The U.S. Patent and Trademark Office (USPTO) is tarred with this pearl of wisdom and has never been able to quite wash its hands of it (even though its actual birth is lost somewhere in the 1800s). Now, in the latest opportunity to show that it is not the philistine in the room, the USPTO is soliciting and considering the public’s views on patenting artificial intelligence inventions. Not inventions about how to design artificial intelligence. Not inventions about how to utilize artificial intelligence. But inventions that artificial intelligence has constructed. Of course, true AI doesn’t exist yet, and the inquiry more correctly relates to pseudo-AI inventions, but that doesn’t water down its importance and the direction of thought on the subject.
This evaluation is in stark contrast to the more vanilla (but necessary) efforts of the USPTO and the other major global patent offices to develop their own AI tools to help examiners analyze patent and trademark applications. Broadly, the USPTO is looking at how AI inventions could fit under current patent regulations and whether new forms of intellectual property protection are otherwise required.
Other intellectual property and related issues, like copyright and data protection, face similar uncertainties. The crux of the dilemma is whether or not to consider an AI as an inventor, author, owner, etc. – that is, equal to a human person.
“AI tools helps examiners to analyze patent and trademark applications”
Without tripping into a rabbit hole of the ethics regarding AI, robots, etc., it is clear there is a real problem to be addressed. Machine learning algorithms are now indispensable tools, being used across industries and on multitudinous dataset types. But as they venture into creativity and design, the connection between the AI researchers and the AI results are likely to get so tenuous as to wonder if the tool has overtaken the master. If this ambiguity is not resolved, current laws and regulations (in the U.S. and abroad) could lead to innovations and inventions that will not be protected, putting research investment at risk. And the challenges are already occurring.
But there is too much economic investment and growth at stake for no fix. To be sure, laws are intended to maintain the relations between human persons, although their goals may be to encourage abstract principles. So, the protection of creative works and inventions in the U.S. stems directly from the U.S. Constitution (“The Congress shall have Power… To promote the Progress of Science and useful Arts…”), but the rights granted indisputably belong to human persons.
Yet, if AI will be part of how we invent, there is room to create a new regime of IP protection of those AI results without the de-humanization of our laws and, more importantly, our human activities. Some may frame the problem as a high-minded fight against outdated ideas and for insuring the advancement of the arts and science. Ultimately, however, this is actually about economic control of innovations, inventions, and, of course, money flows.