Anthropic confirms it runs a wet lab to test its AI biology models

Anthropic has confirmed it operates a wet lab, a facility for hands-on experiments, in the Bay Area, where it uses its AI models to test biology research in real life. The company told TechCrunch that AI models run physical experiments there, moving beyond software into laboratory work.

Eric Kauderer-Abrams, Anthropic’s head of life sciences, told Reuters that lab work remains essential to biology research. “We believe that to do biology, the final test is still, and will be for a while, in real lab work,” he said. “We absolutely are doing that today.” He added that the lab functions like most biotech labs, with Anthropic running some research itself and working with external partners on other projects.

The move follows Anthropic’s April acquisition of Coefficient Bio, a stealth AI biotech startup. Anthropic declined to detail what the lab is working on but said its main focus is fundamental biology rather than drug discovery. The company has avoided products that could be seen as competing with pharmaceutical customers, including Novo Nordisk, with which it recently announced a joint drug discovery deal. Anthropic has faced backlash before for launching products that overlapped with its own customers’ businesses.

This week Anthropic also launched a Life Sciences Verification Program, which gives vetted biology researchers access to its most capable models. The company has published research reports on using AI to accelerate protein design and on applying its models to biomolecular modeling.

A lab from a company that warns about AI risk

The wet lab news sits alongside Anthropic’s own warnings about AI danger. Researcher Jacob Coxon resigned this year, saying “the people building AI earnestly believe that it could kill us all by the end of the decade.” Anthropic’s alignment lead has put the odds of AI causing human extinction within the next decade at above 10%. CEO Dario Amodei published a post last weekend urging the AI industry to slow down and adopt self-regulation, and has repeatedly named bioterrorism as one of AI’s biggest risks. Amodei has also said he believes AI could cure most major diseases within five to ten years.

The contrast has drawn public mockery. Investor Chamath Palihapitiya wrote on X: “The group behind such hits as: ‘We’re All Going To Die’ and ‘Regulate Me Now’ are building a wet lab in SF. I do not recommend this.”

Anthropic has not said what specific experiments the lab is running or which external partners it works with. Whether the company discloses more detail as the Life Sciences Verification Program expands, and how it draws the line between fundamental research and drug discovery work that could compete with partners like Novo Nordisk, are the next things to watch.