Anthropic's Claude model designs proteins that target scorpion toxins
Anthropic's internal research Claude model was asked to design small proteins that lock onto scorpion neurotoxins.
By Standard Topic. AI-generated from the reporting listed under Coverage. Published . Updated .
This story is no longer being updated. It reflects reporting up to .
What happened
Anthropic's internal research Claude model was asked to design small proteins that lock onto scorpion neurotoxins. Citrini Research said it had spent three months using the model to design proteins meant to work in a real lab. Of the designs tested in the lab, 17 met the criteria, and the two strongest bind to Cn2.
Why it matters
Their real-world effectiveness is still an open question. The results so far come from computer-generated designs tested in the lab.
Timeline
- Oct 20, 8:59 AM ET: Citrini Research published an account of its three months of work with the model on protein design.
- Oct 20, 9:01 AM ET: Anthropic published its own account of the model designing binders against scorpion neurotoxins.
- Oct 21, 4:37 AM ET: Lab testing results were reported, showing 17 of 211 tested designs met the criteria.
Coverage
- Dealroom.co | Citrini Research let Anthropic's Claude design scorpion antivenom: 211 proteins tested, 17 binders, then months of waiting on the lab (app.dealroom.co)
- Claude Science tackles deadly scorpion stings by designing protein binders from scratch (the-decoder.com)
- How Claude designed protein binders against scorpion neurotoxins (anthropic.com)
More from Technology & AI
- China puts humanoid robots to work serving food and sorting parcels
- Newsom proposes sweeping AI safety rules for Washington
- Google releases Gemini 4 Neon AI model with million-token context
- Microsoft and Intel widen their cloud chip partnership
- Google DeepMind rolls out Genie 4 to select subscribers
- AI helped AMD shorten chip design work, OpenAI says