According to a new report from the scientific journal Nature, numerous researchers have successfully used proprietary commercial AI tools to create antibodies that have the properties of antibody drugs – drugs used in the treatment of cancers, autoimmune disorders, infections (including COVID-19), and inflammatory conditions.
With the speed of progress these AI-designed drugs are seeing, scientists estimate that clinical trials could begin soon, meaning that in a few years AI-designed drugs could undergo FDA review for approval, possibly entering the market.
In early 2025, the FDA issued draft guidance to provide recommendations on the use of AI to support regulatory decisions about drug safety, efficacy, and quality. While policy is still forming, the FDA recognizes the increased use of AI throughout the drug product life cycle and across a range of therapeutic areas, and has displayed a willingness to collaborate with those using AI for drug development.
While AI models used in drug development may be effective in optimizing certain design elements, many critics have pointed to the “black box” problem, where the decision-making process for why an AI model shaped a drug the way it did, is vague at best.
This decision-making process has historically been significant in drug development, and by extension, the regulatory framework surrounding pharmacy management. Drugs approved by the FDA are assumed to have gone through effective safety and efficacy review, and when clinically appropriate, can be utilized for the treatment of relevant conditions.
However, AI-designed drugs may prove to be a different story. How they could impact drug policy in the near future remains to be determined.
