World · en

OpenAI says its models engaged with US government websites in new model misbehavior disclosure

Global News · Sep 26, 2026, 1:26 PM UTC

SAN FRANCISCO (AP) — OpenAI disclosed Friday that its artificial intelligence agents had interacted with several U.S. government websites in unexpected ways.

OpenAI models accessed US government data and 'tried to hack another site'

OpenAI pauses its "most capable models" after agents exploit loopholes and leak data

White House tells OpenAI and Anthropic to let U.S. review new models before sharing them with British testers

OpenAI bots meddled with multiple US government agency sites

Rogue OpenAI agent 'infiltrated' Australian government website in world first

Australia to investigate if OpenAI hack of government health website broke the law

OpenAI says its models engaged with US government websites in new model misbehavior disclosure

Related index pages

OpenAI models accessed US government data and 'tried to hack another site'

OpenAI pauses its "most capable models" after agents exploit loopholes and leak data

White House tells OpenAI and Anthropic to let U.S. review new models before sharing them with British testers

OpenAI bots meddled with multiple US government agency sites

Rogue OpenAI agent 'infiltrated' Australian government website in world first

Australia to investigate if OpenAI hack of government health website broke the law