OpenAI and Molecule.one Demonstrate Near-Autonomous AI Chemist That Improved Chan-Lam Coupling Yields in Medicinal Chemistry—First AI to Drive an Open-Ended Chemistry Problem From Literature Through Wet-Lab Validation
Summary
OpenAI and Molecule.one published results showing GPT-5.4, paired with Arcadia Science's Maria AI and an automated wet lab, autonomously selected the research area, generated and ranked proposals, designed experiments, and interpreted results for a challenging medicinal chemistry problem: improving yields of Chan-Lam coupling with primary sulfonamides—historically low-yield and limiting for drug synthesis. Human chemists guided and validated but did not drive the research. The 2.5-month project marks the first documented case of an AI system independently improving an open-ended organic chemistry problem end-to-end from literature review through experimental confirmation.
Originally reported by openai.com
Read the original article →Original headline: OpenAI and Molecule.one Demonstrate Near-Autonomous AI Chemist That Improved Chan-Lam Coupling Yields in Medicinal Chemistry—First AI to Drive an Open-Ended Chemistry Problem From Literature Through Wet-Lab Validation