Anthropic AI Models Misused for Biological Weapons Research

• •Anthropic's AI models were targeted for misuse in biological weapons research.
• •The company has blocked multiple accounts and implemented stricter safeguards.
• •A former employee raised alarms about the potential dangers of AI technology.

Anthropic reported attempts by various malicious actors to misuse its AI models for developing biological weapons and conducting cyber operations. The threat intelligence report, published on September 10, 2026, detailed multiple cases of misuse, including a prompt for grant applications related to gain-of-function research on the chikungunya virus. The report highlighted the involvement of state groups, spyware vendors, and politically motivated individuals from countries like Russia and China. Anthropic has blocked these accounts and implemented stricter safeguards on its AI models to prevent further misuse. The report follows the resignation of a former employee who warned about the potential for AI to cause human extinction by 2030. The company emphasized the need for responsible AI development and increased regulatory scrutiny. The misuse cases included not only biological research but also cyber espionage and propaganda efforts.