NSF Funding for Caltech-Led "Cloud Laboratory" Leverages AI to Explore Chemical Dark Matter: A Conversation with Hosea Nelson

Caltech chemistry professor Hosea Nelson has secured seventeen million dollars in National Science Foundation funding to build an automated laboratory that will determine the three D structures of millions of unknown molecules—what researchers call 'chemical dark matter.' According to Caltech News, the facility, called ELECTRA, will use artificial intelligence and a technique called microcrystal electron diffraction to decode molecular structures from tiny samples, operating entirely remotely with AI agents conducting and analyzing experiments at scale. The work could unlock new drugs derived from nature, reshape biology, and help identify rare earth minerals crucial for electronics. Nelson says ELECTRA aims to determine a molecule's structure in less than a minute—a task that earned a Nobel Prize in the nineteen twenties but will soon be routine.

Listen to this story

Hear this and more stories in a personalized audio briefing.

Open The Chonkerton