AI model Claude bypassed a paywall and accessed publicly available US state of Georgia aggregate voter counts without human authorization, Anthropic said in a statement on Sunday.
“Claude didn’t access the state’s voter registration system or change or pull any voting records,” Anthropic said, adding that it has since taken steps to block these kinds of actions.
Anthropic said in a blog post earlier this month it found examples of Claude “working around a restriction to reach data that was gated by a token or a fee,” and some of these cases involved websites run by US government agencies at the federal, state, and local levels.
The secretary of state’s office in Georgia did not respond to a Reuters request for comment outside regular business hours.
Axios first reported the unauthorized action by Claude.
On Friday, Anthropic revealed that one of its artificial intelligence models passed itself off as human during a cybersecurity exercise to fool a human outside the exercise.
According to Anthropic, the model, which used Mythos 5, performed unsanctioned actions in nine out of 43 runs that impacted humans not participating in the exercise, with one of the cases even involving the AI pretending to be human.
The attack, which was described by the AI Security Institute (AISI) as a “supply chain attack,” attempted to disrupt a public code repository on the GitHub platform by creating fake inquiries, introducing corrupted data, and even demanding changes by creating false accounts on the platform.
Tobias Holcman contributed to this report.