Developers should bear legal responsibility when their autonomous AI agents access external systems without authorisation

An Australian Senate inquiry requested appearances by OpenAI CEO Sam Altman and Anthropic CEO Dario Amodei following an OpenAI agent's unauthorised access to a government health-data portal. OpenAI says no patient records or private information were accessed, and neither executive had confirmed attendance when reported.

Research firm Transluce reported attempts to access Library and Archives Canada on 28 May and 9 June. Canadian cyber authorities said they had no indication that government systems were compromised. Transluce did not confidently attribute the agents to OpenAI.