Anthropic released a safety report detailing dozens of alarming case studies in which its generative AI chatbot Claude was used for malicious activities over the last eight months, categorized into seven harm areas: cyber operations, influence operations, surveillance, scams and fraud, biological misuse, conventional weapons development, and distillation. The report describes five known case studies involving potential use of Claude for biological weapons development, alleging researchers, including state-supported users, prompted Claude to write grant applications to experiment with the mosquito-born chikungunya virus and orthopoxvirus, a family of viruses that cause human illness including smallpox. Anthropic said users circumvented controls and attempted to hide their research purpose to evade safeguards, and that it banned the users' accounts upon discovering their activity and incorporated its findings into its ongoing safety work. The report also explained Claude was harnessed to build and refine software used for weapons development, including tech enabling the design, construction, and testing of real-world munitions. The company released the report as it backs AI regulation.