The pharmaceutical industry, which is actively integrating artificial intelligence into the process of developing new drugs, has found itself in a difficult legal situation. U.S. legislation strictly requires that only human inventors be named in patent applications, which creates serious obstacles for companies where neural networks play a key role.
The case of Insilico Medicine is telling: the company stated that its generative AI platform synthesized a promising compound for combating pulmonary fibrosis. However, the artificial intelligence does not appear in the patent application at all—five people are listed as inventors, including CEO Alex Zhavoronkov. This is not a coincidence but a forced measure dictated by judicial practice.
The key precedent was the DABUS case, reviewed by an appellate court in Washington in 2022. The court confirmed that the term "individual" in U.S. legislation is interpreted exclusively as a human being, even when it concerns an improved food container designed by AI. This decision effectively closed the door to directly recognizing neural networks as inventors.
Now disputes are shifting to the assessment of human contribution. What becomes decisive is not the mere fact of using AI, but how significant human involvement was in the process. Ryan Abbott, a partner at the law firm Brown, Neri, Smith & Khan, warns: if inaccuracies surface in the documentation, a wave of patent challenges is inevitable. His position is categorical: "If I asked Claude to cure cancer and it did so, it would be inappropriate to claim that this was my achievement."
Notably, the regulators' position has changed multiple times. The U.S. Patent and Trademark Office (USPTO) initially issued detailed guidelines on criteria for co-authorship when using 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.
Sarah Korman, chief legal officer and head of development at Isomorphic Labs, believes that current legislation is outdated. As the role of AI grows, it will have to be revised, but for now companies are forced to artificially keep humans in the development chain—documenting their involvement in the synthesis, modification, and testing of molecules to meet formal requirements.
Recall that in July, the AI startup Anthropic launched Claude Science, an environment for scientists, and announced plans to develop drugs for "neglected" diseases, which will only increase pressure on the legal system.
My analysis: The situation resembles the legal vacuum of the early internet era, when legislation lagged behind technology. While courts and regulators seek a balance, innovation in pharmaceuticals risks slowing down. Companies that can legally formalize human involvement in AI developments will gain a competitive advantage, but a long-term solution will inevitably require legislative changes.