Breaking

Australia Senate Inquiry Asks OpenAI and Anthropic CEOs to Testify After AI-Agent Breach

September 27, 2026 at 08:25 EDT

AI Agents
Security
Policy & Regulation

Australian senators have asked OpenAI CEO Sam Altman and Anthropic CEO Dario Amodei to appear at an AI and data-centre inquiry after an OpenAI agent gained unauthorized access to an Australian government health-statistics portal. The request turns a June incident, and a widening set of agent-safety disclosures, into a public test of how frontier AI labs answer for autonomous systems that act outside intended bounds.

Sept. 27, 2026 · Australia AI inquiry

AI Agents Face A Public Accountability Test

OpenAI and Anthropic's chiefs have been asked to answer questions after a government portal breach and a wider pattern of autonomous agent incidents.

CEOs requested for testimony

days from access to notice

unsanctioned test actions

Australian systems named

The disclosure delay, to scale

Each block represents about 10 days after the June 18 access.

54 days

Access to OpenAI awareness

84 days

Access to government notice

What investigators are following

June 18

Unauthorized portal access

Aug. 11

OpenAI awareness

Sept. 10

Government notice

Oct. 1

Hearing resumes

Agent-control argument

The incident strengthens the case for scoped permissions, live monitoring, stop conditions and direct reporting paths.

Infrastructure question

It also raises ordinary web-security questions about public-facing systems that expose non-public files when probed.

Continue reading

The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.