As the debate intensifies over whether artificial intelligence (AI) should be recognized as an inventor, the heuristic we identified in February between autonomy and automaton can fruitfully be applied.
Some technologists, policymakers, and philosophers argue that AI is (or soon will be) an authentically autonomous entity capable of cognition and creativity. Others contend it’s no more than an extension—and a servant—of its designers.
It’s also worth pointing out that both the autonomy and automaton viewpoints have positive and negative variants. Positive automaton-ers believe we can improve the outputs of AI by ensuring our inputs are fair and equitable; negative automaton-ers mock AI’s inability to live up to the abilities of the human brain. Positive autonomists wax enthusiastic over the boundless possibilities of, say, future versions of ChatGPT; negative autonomists wax terrified over the likelihood that machines will one day become our overlords.
So how does the automaton-autonomy distinction map onto the ongoing discussion of AI inventions?
Recall that prolific inventor Stephen L. Thaler devised a machine he called DABUS, or a Device for the Autonomous Bootstrapping of Unified Sentience. (You can likely guess which quadrant Thaler falls into.) As we’ve previously explored, Thaler filed patent applications around the world in DABUS’s name for the AI’s inventions, including a food container that robots can grasp and an emergency warning light.
Thus far, only South Africa has agreed with Thaler that DABUS should be named as an inventor while most other countries—including the United States, the EU, and Australia—have denied the DABUS patent applications.
As we previously discussed, Robert Jehan, one of the patent attorneys advocating for DABUS, believes that we are rapidly approaching “technology that an associated person (such as the creator of the system, the original programmer, the user of the machine, and so on) cannot reasonably say he or she truly contributed to in terms of having taken part in inventing.” True to form, Jehan is emphasizing the “autonomous” aspect of DABUS.
Similarly, Reuven Mouallem, a patent practitioner who has filed patent applications on behalf of DABUS in Israel and the United States, explained to me a few years ago that the machine’s design reflects “a quality most akin to sentience, a chaining of conceptual hives resembling the formation of a series of memories.”
In contrast, Huw Jones, the deputy director of the United Kingdom Intellectual Property Office (UKIPO), which denied DABUS’s initial application, asserted that “DABUS has no rights to its inventions and cannot enter into any contract to assign its right to apply for a patent to the applicant.” Jones also found that “an AI machine is unlikely to be motivated to innovate by the prospect of obtaining patent protection.”
Likewise, the European Patent Office held that “AI systems or machines can be neither employed nor can they transfer any rights to a successor in title.” According to this UK-EU view, DABUS is an automaton programmed by Thaler without independent motivations and certainly without rights.
This debate came to the fore earlier this month when the UK Supreme Court considered Thaler’s appeal of the UKIPO decision. Jehan argued that “DABUS has a mechanism to identify itself the novelty and salience of the present invention.” (Emphasis added.)
But during the hearing, one lord justice expressed skepticism of the “assumption that the AI system is the creator,” while another characterized the DABUS team as “intransigent” and having “run into a brick wall.” Likewise, counsel for the UKIPO emphasized the “moral right, the right to get credit for what you’ve done, a right which, in my submission, can only vest in a person.” Robots need not apply.
How the UK Supreme Court ultimately decides the DABUS issue remains to be seen, but the automaton-autonomy distinction lives on.
The post Debate over AI Patenting Reflects Autonomy-Automaton Distinction appeared first on American Enterprise Institute – AEI.