- Published on
Anthropic details bad actors’ efforts to misuse its AI for bioweapons
- Authors
- Name
- DeepHorus
Anthropic has revealed that criminals, state-sponsored groups, spyware vendors, scientists and propagandists have attempted to use its artificial intelligence models to design missiles and bombs, create deadly pathogens and surveil dissidents. The threat intelligence report, published on Thursday, spans 154 pages and details what the company calls the most notable and novel misuse identified to date. The disclosure comes two days after a former employee quit, warning that Anthropic's models could cause human extinction by 2030.
The company said it published the findings because it has a responsibility to disclose malicious misuse of its services. In particular, Anthropic highlighted five case studies of scientists using its AI models in biological research. In those examples, researchers circumvented safeguards meant to prevent 'unsupported regions' and worked to hide the purpose of their work.
Anthropic said it banned the accounts involved, but did not reveal the names of research institutions or countries, citing uncertainty about the researchers' intent. The report also detailed dozens of other threat-actor cases. These included Russian espionage and 'smash-and-grab' cyberhacks, a China-based surveillance program targeting Uyghurs in Syria, and propaganda campaigns in Russia, Malaysia, Iran and Bangladesh.
Additionally, Anthropic's Claude AI model was used in Yemen, China and Russia to develop weapons. The company described biological misuse as one of the most serious risks of frontier AI models. 'Without the correct safeguards, such capabilities could have catastrophic consequences,' Anthropic wrote.
The report underscores the growing security concerns surrounding advanced AI. Anthropic emphasized that these cases are not typical misuse, but rather notable and novel threat activity. The company's decision to publish the report reflects a broader industry debate about transparency and safety.
Going forward, the findings are likely to intensify calls for stronger AI safeguards and regulation. Anthropic has not disclosed whether it has notified affected institutions or governments. The company maintains that disclosing malicious misuse is part of its responsibility.
