From Engadget · Mariella Moon · · 1m 30s
Anthropic opens biology lab for AI experiments
The AI company is building physical lab capacity, but won't say what it's actually researching.
In brief
The AI company is building physical lab capacity, but won't say what it's actually researching. Anthropic is moving beyond software into wet lab biology, combining AI with hands-on experimentation, though the company faces trust concerns around AI safety. Originally reported by Engadget.
Anthropic opens biology lab
The AI company has established a research lab in the San Francisco Bay Area to conduct physical experiments. Head of Life Sciences Eric Kauderer-Abrams confirmed the lab exists and said real lab work remains essential for biology.
“We believe that to do biology, the final test is still and will be for a while in real lab work.”
Drug discovery program launched
Anthropic launched Claude Science in June to research drugs that pharmaceutical companies avoid because they're not profitable. A spokesperson said the new lab wasn't created specifically for drug discovery alone.
AI could accelerate drug development

Kauderer-Abrams believes AI could speed up discovery of treatments for conditions previously thought too difficult to treat, including complex and targeted therapies.
Teaching Claude to control robots
Anthropic is exploring ways to make Claude control robots to carry out scientific experiments. Kauderer-Abrams said the company is in very early stages of automating lab work with AI.
Human oversight still essential
Kauderer-Abrams emphasized that human oversight remains critical for safety as AI automates lab work. Anthropic has not yet run any clinical trials.
Trust problem looms large

Anthropic faces public skepticism like other AI companies. The firm recently stopped multiple scientists from using its models to create potential biological weapons.
Here's the gist
Anthropic is moving beyond software into wet lab biology, combining AI with hands-on experimentation, though the company faces trust concerns around AI safety.



