AI Scientist Autonomously Generates and Validates Biological Discoveries

Researchers at Chalmers University of Technology have developed an AI system capable of autonomously generating and validating new biological discoveries through laboratory experiments on brewer's yeast. By combining large language models, automated reasoning, and laboratory automation, the team created a closed-loop AI laboratory.

The AI was provided with scientific knowledge, including the yeast's genome, metabolism, and previous studies. Subsequently, the AI could identify promising biological questions, recommend experiments to test them, and evaluate the outcomes. According to the research, the AI can actively generate new scientific knowledge rather than solely supporting human decisions.

"AI scientists will collaborate with human scientists to accelerate discoveries across biology, medicine, and biotechnology," stated Professor Ross King, the study's senior author. He compared the development to self-driving cars, where AI processes information and takes action.

The researchers emphasize that AI systems will augment, not replace, human scientists. These systems can handle routine cycles of hypothesis generation and experimental testing, thereby accelerating the exploration of complex scientific questions and optimizing laboratory resources. Human scientists retain their essential role in defining research priorities, interpreting scientific significance, and ensuring ethical oversight.