Artificial Intelligence: AI-Assisted/Generated Innovation - 4 Key Takeways

Kilpatrick Townsend Partner Rich Christiansen recently was a panelist discussing the role of AI as an innovator and whether AI-assisted/generated innovation should be protectable. Key takeaways from the discussion include:

1. AI is already the sole innovator behind many patent applications, and AI’s role in innovation is only increasing. The level of AI use in innovation varies across different industries, but there are several industries where AI is the innovator behind a significant part, if not the majority of new patent applications. These industries currently identify individuals that were at least nominally involved with the AI as inventors.
2. There is no perfect solution to the problems posed by AI as an innovator. Allowing AI to be identified as an inventor may lead to a deluge of patents and patent applications as the relative cost of innovation decreases as compared to the exclusionary benefit of patent protection. This could effectively exclude new entrants to the market and consolidate market share amongst a very small number of large companies. On the flip side, preventing AI to be identified as an innovator will lead to the misattribution of innovation and build a lie into our system. It may also decrease the incentive to developing AI technologies.
3. The rise of AI as an innovator may change application of obviousness/inventive step. The legal concepts of obviousness and inventive step are currently based on the actions of an imaginary person of ordinary skill in the art. In the future, AI may provide us with a digital- incarnation of that person of skill in the art. Obviousness challenges may then revolve around the type of AI, the training set, and how the AI is trained.
4. Unless the potential risks and challenges of AI are proactively addressed, we will likely end-up with layers of legal-fictions created by courts in an attempt to shoe-horn a new reality into the structures of a passing epoch.