Notice: Anthropic Disrupts Attempts to Misuse Its Model for Biological Weapons Work

Anthropic has said it identified and disrupted attempts to use its model for "malicious activity" that could support the development of biological weapons, according to the company's latest threat intelligence report. The findings are the latest in a growing series of warnings from researchers and industry insiders about the potential risks of advanced technology.

What the report found

The report highlighted five case studies of actors using the company's models in ways that could support biological weapons development, and noted six cases where the technology was used to develop software for conventional weapons, including firearms, missiles and armed drones. Over the past eight months, cases of concern ranged from fake dating apps and hotel Wi-Fi scams to surveillance built to identify dissidents. Misuse of the company's models was disrupted between December 2025 and August 2026, the report said, and the company accused some foreign firms of trying to replicate its capabilities.

Broader industry warnings

Such reports have become a regular feature across the industry as firms seek to show how they identify and disrupt misuse. Google wrote this week that a person had attempted to use its tool to obtain a step-by-step technical guide for synthesising weaponised biological agents. The warnings have prompted calls to action: a US senator has demanded a pause on advanced development and a ban on superintelligence, while the US president has so far rejected such fears. A chief scientist at a rival firm called for voluntary slowdowns until safeguards are set, and an open letter urged a new multinational treaty for safe development.

Safeguards and next steps

Anthropic said biological misuse is one of the most serious risks of frontier models and that, without the correct safeguards, such capabilities could have catastrophic consequences. It noted that the same information that could be used to develop a biological weapon could also be used to develop a vaccine or a cure. The company said it had incorporated its findings into its processes to better prevent, detect and disrupt such activities, and had shared intelligence with authorities and industry partners where appropriate.

• Five case studies of misuse that could support biological weapons development
• Six cases of software development for conventional weapons
• Misuse disrupted between December 2025 and August 2026
• Findings incorporated into company processes and shared with authorities
• Part of a wider wave of industry warnings and calls for safeguards

Company details

Anthropic is an artificial intelligence safety and research company. Media enquiries are handled by the company's communications team. Website: anthropic.com.