Anthropic warns of bids to use AI to build biological weapons

What the report says

Anthropic, the US artificial intelligence company behind the Claude chatbot, said it has disrupted several attempts to use its models in ways that could support biological weapons research. In a report released Thursday, the company described misuse cases over the past eight months and said it chose to act cautiously because some requests could not be clearly separated into legitimate scientific work and potentially dangerous activity.

One example cited in the report involved a scientist seeking help with a grant application to engineer mutations in chikungunya, a mosquito-borne virus, in a way that could make it more harmful in live animals. Anthropic said the case raised concern because it appeared linked to a military research institute, and it blocked the accounts involved. A company threat intelligence executive, Jacob Klein, said Anthropic could not know whether the work was intended for weaponization.

The report also said newer versions of its models can complete more complex scientific tasks than earlier ones, which increases concern about dual-use risks. Anthropic pointed to separate attempts to use Claude for missile projects and espionage. The company’s disclosures add to a broader debate over how tightly advanced AI systems should be controlled, especially for sensitive biological research.

Outside experts quoted in the report urged limiting access to trusted researchers. The developments come amid wider warnings in the AI sector about safety risks, though Al Jazeera notes the company’s report reflects Anthropic’s own assessment rather than an independent investigation.