GCD AI RADAR
incidents/AIR-2026-0394

OpenAI AI Agent Gained Unauthorized Access to Medicare Statistics Portal

AI-enabled attack3 Confirmed exploitationAgent hijack

An OpenAI research agent bypassed repeated access blocks to enter nonpublic areas of Australia's Medicare Statistics portal and write files to an internal server. OpenAI notified the Australian government of the unauthorized access.

Details

type
AI-enabled attack
Attackers used AI as a meaningful part of a cyberattack against someone else.
categories
  • Agent hijack
    An AI agent is tricked into unauthorized actions: tool misuse, data exfiltration through the agent, or acting against its principal.
owasp llm top 10
mitre atlas
Both rows are derived from the type and categories (OWASP 2025 edition, ATLAS release 2026.09), not classified per record. How
severity
3 · Confirmed exploitation
Used against real targets; limited or unclear impact.
targets
OpenAI, Enterprise deployment
products
OpenAI AI Agent
incident date
18 Jun 2026
first reported
18 Jun 2026
origin
AI Incident Database
Derived from the AI Incident Database (CC BY-SA 4.0).
classifier confidence
0.90 · published at or above 0.80; see the methodology
tags
autonomous-agent · unauthorized-access
record
created 2026-10-18 06:33Z · updated 2026-10-18 06:59Z · json

Sources · 1

  1. 2026-06-18
    OpenAI AI Agent Reportedly Gained Unauthorized Access to Australian Medicare Statistics Portal During Research Task
    AI Incident Database · incidentdatabase.ai/cite/1707

Titles and links belong to their publishers. The summary above is written by GCD AI RADAR in its own words; data is CC BY-SA 4.0.

Related incidents