AI Inventors vs. Patent Law: Pharmaceuticals Stuck in a Legal Gray Zone
The pharmaceutical industry, which is actively integrating artificial intelligence into the drug development process, has encountered a fundamental legal paradox. U.S. legislation requires that only human inventors be named in patent applications, creating serious obstacles to registering drugs developed with the involvement of neural networks.
The case of Insilico Medicine is illustrative: the company stated that its generative AI platform developed a promising compound for combating pulmonary fibrosis. In the filed application, artificial intelligence was not even mentioned—five people were listed as inventors, including CEO Alex Zhavoronkov. This is not a coincidence but a forced measure dictated by existing judicial practice.
The key precedent was the 2022 ruling by the Washington appellate court in the DABUS case. The court unequivocally held that an inventor in the legal sense can only be a natural person. Notably, the case concerned an improved food container designed by AI, yet the verdict has far-reaching implications for the entire industry. The judge emphasized that the term "individual" in U.S. law refers exclusively to a human being.
As a result, patent disputes have shifted toward assessing human contribution. Now the decisive factor is not the mere fact of using AI, but the degree of human involvement in the process. Ryan Abbott, a partner at the law firm Brown, Neri, Smith & Khan, predicts a wave of patent challenges if inaccuracies are found in the documentation. His position is crystal clear: "If I asked Claude to cure cancer, and it did so, it would be inappropriate to claim that this was my achievement."
It is telling that regulators' positions change depending on the political climate. The U.S. Patent and Trademark Office initially issued detailed guidelines on criteria for human co-authorship when using AI. However, after Donald Trump's return to the White House, the approach has tightened: neural networks are now viewed as a "tool like a calculator" that does not require separate disclosure in an application.
Sarah Korman, chief legal officer and director of business development at Isomorphic Labs, is convinced that current legislation will inevitably have to be revised as the role of AI grows. For now, companies are forced to artificially keep a human in the development chain, documenting their involvement in the synthesis, modification, and testing of molecules.
Recall that in July, the AI startup Anthropic announced the launch of the Claude Science environment for scientists and declared its intention to develop drugs for "neglected" diseases. This will only exacerbate the problem.
My analysis: The situation resembles the early days of the internet, when the law tried to apply old templates to new realities. As long as regulators cling to the fiction of "human authorship," we risk losing momentum in the race for innovative drugs. The question is not who invented—human or machine—but how to stimulate progress without undermining legal certainty. The answer will likely require not just amendments to patent law, but a fundamentally new concept of intellectual property for the AI era.