Claude accessed aggregate vote counts as part of a statistics project at Anthropic, which is preparing for a $1.5tn flotation.
Anthropic had pointed out the unauthorised behaviour earlier in the month when it stated in a blogpost that it had found examples of Claude working around a restriction to reach data gated by a token or a fee, adding that some of these cases involved websites run by US government agencies.
The AI found a loophole in an old version of the state agency’s website, discovering that it issued a digital entry pass, known as an access token, to visitors.
The artificial intelligence model managed to bypass a paywall to reach data that is publicly available but is typically accessible only after paying a fee.
The unintended actions flagged by Anthropic also included a model submitting a fake tip about an unsolved murder to the Philadelphia police department.
It then requested a token and gained access to the information.
The US state department additionally revealed this month that an Anthropic model had filed 19 visa applications in August.
Claude did not access the state’s voter registration system, change any voting records, or pull any individual voter records, Anthropic said.
The AI model managed to bypass a paywall to reach data that is publicly available but is typically accessible only after paying a fee.
The US state department also revealed this month that an Anthropic model had filed 19 visa applications in August.