Written by Arbitrage • 2022-10-05 00:00:00
According to movies, the first robot law is that "a robot may not injure a human being. Or through inaction allow a human being to come to harm." In real life? That's more of a slippery slope.
Does harm mean physical or mental/emotional harm, or some level of both? And if it is some blend of not doing physical or mental/emotional harm, we are caught in an ethical web that the new Artificial Intelligence (AI) cannot escape. For example, did the AI that generated a work of art that won a competition do no harm? Or did it in fact harm trained artists by putting together a work in less than 10 minutes that would take them weeks, if not months to produce in other mediums? Which brings us to today's blog topic - creations made by AIs cannot be patented.
That's right - an AI cannot be a patent inventor, at least not in the US and most other countries. So far, South Africa has granted patents to an AI. Well, one specific AI, generated by inventor Stephen Thaler - his AI the Device for the Autonomous Bootstrapping of Unified Sentience (DABUS) has been denied patents in almost every country he filed for them on his AI's behalf. He argues that the inventions of the machine should be attributed to the machine and not him, as he did not partake in the development of its inventions and is simply its creator.
Patent boards don't quite see it that way. In order to file a patent, it must be attributed to an actual human being as the inventor. The United States Patent and Trademark Office (USPTO) agrees with that statement and of course did not allow the patent filed for DABUS as US patent legislations currently limits the definition of "inventor" to a human being. Of course, wanting to "do right" by his invention, Thaler tried to appeal the USPTO's decision in a district court in Virginia, which sided with the USPTO. He then tried to appeal to the US Court of Appeals for the Federal Circuit in August of this year, and they too sided with the USPTO, stating that "there is no ambiguity: the Patent Act requires that investors must be natural persons; that is, human beings." Thaler, of course, refuses to quit and will petition for review with the US Supreme court next in an effort to have his AI recognized as an inventor.
In the event that the AI is recognized as an inventor, what other things can AI take credit for that had limited to no human interaction in creating? What happens to AI generated images, music, and innovation in the arts? We will be watching this case to see if Thaler wins, and his AI can be listed as an inventor for a patent because the can of worms that opens will be interesting to see live.
For the latest news and updates, don't miss our podcast at https://www.arbitragetrade.com/podcasts/