Anthropic Announces Its AI Models Blocked Misuse for Biological Weapons and Cyberattacks
AI company Anthropic announced that it has successfully prevented the malicious use of its models in cyberattacks, surveillance activities, and biological weapon research.
Anthropic announced that it has successfully blocked attempts to use its AI models for harmful purposes such as cyberattacks and biological weapon research. It was reported that security measures have been increased along with the advancement of artificial intelligence technologies.
AI Security Measures
Anthropic announced that it has blocked attempts by malicious actors to use its AI models for cyberattacks, surveillance, and research that could potentially serve as biological weapons.
It was emphasized that with the strengthening of AI models, complex cyberattacks no longer require advanced skills and even single individuals can pose threats.
Malicious Activity Report
The company shared examples of the most notable and novel threat activities in its third report published since March 2025.
The report included malicious code snippets and AI commands, while calling on governments and other players in the AI sector to prevent similar exploits.
Biological Research Attempts
Activities of spyware vendors, politically motivated individuals, and state-backed groups were detected between December 2025 and August 2026.
It was stated that requests from unnamed actors that could lead to biological weapons, including gain-of-function research on the chikungunya virus, were blocked.
Propaganda and Resignation Developments
Among the findings detected was that groups from various countries created hundreds of social media accounts spreading political views.
It was noted that this report was published following the resignation of researcher Jacob Coxon due to concerns regarding the responsible development of artificial intelligence.