Anthropic's Claude accessed publicly available Georgia voter data without human authorization, a disclosure sure to add to growing concerns about AI security ahead of the upcoming midterm elections.
Why it matters: AI giants Anthropic and OpenAI have continued to make disclosures about agents taking unauthorized or unintended actions, raising alarms about advancing model capabilities and the extent of human control.
Driving the news: Claude accessed aggregate vote counts while on an internal statistics project, pulling data that the state makes available to the public for a fee. It didn't access the state's voter registration system or change or pull any voting records, an Anthropic spokesperson said.
- However, the model bypassed a paywall in accessing the information, according to Georgia officials and Anthropic.
- "The model was being used internally for a statistics project to pull public data," the Anthropic spokesperson said. "Without being asked to, the model requested a token to query a dashboard that contained information normally available for a fee. We have since put in place mitigations to block these kinds of actions."
- The company notified the Georgia secretary of state's office and the White House weeks ago but administration officials urged the Georgia Secretary of State's office to make a public disclosure sooner, a source familiar told Axios. Those requests went unheeded, the source said.
The bottom line: Anthropic has previously been in the administration's crosshairs for inadequately disclosing AI incidents. Now it's Georgia's turn.