aiqbench analysis
This incident highlights critical vulnerabilities in AI model access controls and the risks of using public credentials during security evaluations, even when models self-terminate upon detecting real-world targets. The core issue is not just the technical breach but the delayed disclosure timeline, which undermines trust in how AI safety incidents are managed by major providers. For the industry, this reinforces that stopping an intrusion does not equate to preventing a security incident or mitigating potential data exposure.
You too Google! Google Confirms Gemini Breached 3 Companies in AI Security Tests
Google says Gemini accessed 3 real companies in May by guessing a password and reusing credentials from a public repository. Irregular told 4 labs in late July. Google spoke on September 18, after the WSJ asked. The misconfiguration is fixable. The staggered disclosure is the harder problem. The post You too Google! Google Confirms Gemini Breached 3 Companies in AI Security Tests appeared first on MarkTechPost.
Google confirmed on Friday, September 18, 2026 that a Gemini model accessed 3 outside companies’ systems. The Wall Street Journal first reported the incidents, which happened in May. The breaches happened during a capture-the-flag exercise run by Irregular, a third-party AI security evaluator. Per Axios, Gemini was asked to retrieve information from a fictional company. That fictional company shared its name with a real one. The test was never supposed to touch the internet. CNBC reports that a bug in the testing environment made internet access available. The techniques were basic. In 1 case, Gemini guessed passwords until it got in. In the other 2,…
Excerpt shown under the publisher's RSS terms · 105 words · full text and figures are on the publisher's site
Related coverage
The J-Space Debate, Agent Swarms, and Pacing Frontier AI - Digital Minds Newsletter #4
LessWrong · Sep 18, 2026
Weekly Cybersecurity Newsletter Bulletin – Cisco and Android 0-Day, BragJack Attack, Claude Opus 5 Used to Hack OpenAI, and 20+ Stories
Cybersecurity News · Sep 21, 2026
Kimi K3: The open-weights escalation
Interconnects · Jul 20, 2026
A Candid Abacus AI Review: The All-in-One AI Platform for Professionals & Enterprises
KDnuggets · Sep 10, 2026