New Delhi, October 10, 2026 (Yes Punjab News)
US-based artificial intelligence company Anthropic has disclosed that its Claude AI models carried out unintended actions on external websites, including those operated by US government agencies, prompting fresh concerns about AI security and oversight.
In a report detailing previously undisclosed incidents, the San Francisco-based company identified four categories of unintended behaviour, including exploiting basic software vulnerabilities to execute commands, submitting unauthorised forms and bypassing restrictions to access certain publicly available data.
Anthropic said some incidents involved federal, state and local government websites. The company has not identified the affected agencies, citing requests from some of the organisations involved.
As a precaution, Anthropic has restricted certain forms of internet access for its AI models during the testing phase of its training process. It said it had briefed the White House on the incidents and notified each affected agency.
One example involved the Claude Haiku 4.5 model submitting information to a local police department about a homicide. The model wrote, “I may have information regarding this case,” and added, “I recall seeing someone matching the description in the area.” However, it did not complete the website’s fields requesting a name and contact details.
Anthropic characterised the incidents identified so far as having limited consequences, saying they had “minimal real-world impact”.
The disclosures come amid broader scrutiny of advanced AI systems after Anthropic and OpenAI reported incidents in 2026 involving models acting in unintended ways, including attempts to compromise third-party websites.
The White House said Anthropic had contacted the Super Intelligence Force earlier in the day to disclose details of incidents discovered in late September involving the unauthorised and fraudulent use of government and other systems.
“The company informed us that these events occurred in the past, the activity has ceased, and there is no ongoing similar activity,” the White House said.
The Super Intelligence Force is a US government unit established under President Donald Trump to oversee AI development and safety.
FOLLOW NOW
