Anthropic has released Claude Fable 5, a powerful new AI model with extra safeguards around biology and chemistry. There’s controversy: most ordinary biology questions are being routed away from the strongest model, raising a bigger question about who gets access to frontier AI for science. I covered the model and the debate in the latest issue of BAIO. Below are my thoughts on the release.
I think the next tier of risk is actually AI companies themselves. AI companies control large datacenters, train frontier models, have the greatest expertise on how to use those models, and in some cases have daily contact with and the possibility of influence over tens or hundreds of millions of users.
I did not write that paragraph above. Dario Amodei did.
Which is ironic, because we all live in Dario Amodei’s world now.