Anthropic says it blocked misuse of its AI that could have supported biological weapons

Anthropic, an artificial intelligence startup, has reported blocking several instances of malicious use of its AI models, including efforts to support biological weapons. As AI models become more powerful, the potential for misuse increases, Anthropic warns. The company has added stronger safeguards to its latest models to restrict biological research that could be used for weaponization.

Notable misuse cases include attempts to use AI for research on the chikungunya virus, aiming to enhance its transmissibility and immune evasion properties, which could be used to create a more dangerous pathogen. Anthropic has also identified misuse by spyware vendors, politically motivated individuals, and state-sponsored groups.

The report, published amid concerns about the responsible development of AI, urges governments and competitors to identify and prevent similar abuse. The startup emphasizes the need for stronger safeguards as models become more capable, acknowledging the potential risks and responsibilities that come with advanced AI technologies.