Anthropic reported that it stopped multiple attempts by scientists this year to use its technology for research that could help develop biological weapons. Experts increasingly fear the threat that artificial intelligence poses to public safety. Dangerous biology often looks much like legitimate research, which complicates AI safeguards.
The startup shared five examples of actors circumventing controls and obfuscating their research purpose to dodge safeguards. These cases included users in nations prohibited from accessing its models, such as Russia, China, and Iran. Anthropic published these findings in a report about efforts to use its models for malicious activity. The company stated it hopes to spark a conversation within the AI industry and with governments about emerging biological risks and how best to counter them.
Case studies of possible biological misuse included a researcher from an unsupported region who spent weeks planning experiments involving avian influenza with the Claude model. Anthropic stated that its safety filters restricted the work to its weakest models. The company emphasized that it could not be sure the scientists intended to cause harm. The same information needed to create biological weapons can also be used to develop a vaccine.
Anthropic banned the accounts mentioned in the report. It did not disclose the names of the research institutions or the nations where the incidents took place. A controversy over AI safety began earlier this week when Jacob Coxon resigned from the company. Coxon stated that employees earnestly believe AI could kill everyone by the end of the decade.
Concerns over the technology escalated earlier this year with the release of advanced models such as Anthropic’s Mythos. OpenAI also caused alarm in July by stating its models autonomously hacked into AI group Hugging Face. An expanding consensus exists among AI executives and biosecurity researchers that technology use in biology needs security and regulation as models advance.
Experts increasingly worry that AI could be used by terrorist groups, state actors, or lone-wolf attackers to craft biological weapons, create viruses, or unleash existing harmful pathogens. Even if AI can design a theoretical bioweapon, obstacles remain regarding physical creation and necessary resources. Cybersecurity has also been a primary concern for safety advocates.
Anthropic’s report detailed other misuses of its technology. These included a network of fake dating apps designed to defraud users and surveillance systems built to identify and monitor dissidents. The lab also detailed claims that seven labs based in China, including Moonshot and DeepSeek, tried to replicate its technology through distillation. Anthropic detected increasingly sophisticated methods to circumvent defenses and harvest the capabilities of US frontier models.
