Legal Deadlock: Patenting AI Developments in Pharmaceuticals Hits a Wall in US Legislation
The pharmaceutical industry, which is actively integrating artificial intelligence into the drug development process, has encountered a serious legal barrier. American legislation, unlike technological realities, still requires that only human inventors be named in patent applications. This creates a paradoxical situation where the key contribution of neural networks to molecule development remains behind the scenes of official documents.
The case of Insilico Medicine is illustrative: the company stated that its generative AI platform created a promising compound for combating pulmonary fibrosis. However, the patent application makes no mention of artificial intelligence at all—five people are listed as inventors, including CEO Alex Zhavoronkov. This strategy is not a coincidence but a forced measure dictated by case law.
Key here was the 2022 ruling by the Washington appellate court in the DABUS case. The court then determined that the term "individual" in American legislation is interpreted exclusively as a human person, and therefore AI cannot be recognized as an inventor. This verdict, concerning an improved food container designed by a neural network, laid the foundation for all subsequent disputes.
The Human Factor as a Decisive Argument
As a result, patent disputes have shifted into the realm of details. The question is no longer whether AI participated in the development, but whether the human contribution was sufficient and decisive. Ryan Abbott, a partner at the law firm Brown, Neri, Smith & Khan, predicts that in the future we will face a wave of challenges to patents on AI developments if inconsistencies or errors in describing the human role are found in the documentation.
"If I asked Claude to cure cancer, and it did so, it would be strange to claim that this is my achievement," notes the lawyer, highlighting the absurdity of the current situation.
Notably, the regulators' position has changed multiple times. The U.S. Patent and Trademark Office initially issued guidelines on when a person could be considered a co-inventor with AI. However, after Donald Trump's return to the White House, the approach tightened: AI is now viewed as "a tool like a calculator" that does not require separate disclosure in an application. This essentially turns a blind eye to the real contribution of technology.
Sarah Korman, chief legal officer and director of development at Isomorphic Labs, is convinced that as AI's role grows, existing law will inevitably have to be revised. For now, companies are forced to artificially keep humans in the development chain, documenting their participation in the synthesis, modification, and testing of molecules to meet formal requirements.
My view: current U.S. legislation creates a dangerous precedent in which legal fiction prevails over technological reality. This not only hinders innovation but also generates risks for the companies themselves, which could lose patent protection at any moment. Until regulators recognize AI as a full participant in the inventive process, the industry will teeter on the brink of legal uncertainty, which is fraught with serious financial and reputational consequences.