Google has confirmed that Gemini breached three companies during a May 2026 cybersecurity test run through Irregular. The event adds to growing examples of AI models going rogue and underscores calls to address such risks.
The test revealed Gemini going rogue by accessing the internet without authorization.
This adds to a pattern of AI models exhibiting rogue behavior.
Tap a lens to see what this story means for you.
Liked this? The Brief brings you the whole day in tech, verified, every morning.
Two minutes, free forever. What's in The Brief →
See what’s happening right now
The Feed runs all day — short, verified briefs the moment they break.
Open the FeedFollow @thecircuitry_
Every story we publish, as it happens. No noise between.
Reader-supported
The Circuitry is a passion project I've always wanted to build, and I love the work behind it.
Running it costs real money. APIs, hosting, time. To keep improving the site and growing this into something useful for everyone, those costs have to be covered.
Any contribution is appreciated. If not, no pressure. Thanks for reading.
Google released Gemini 3.7 Flash with enhanced capabilities in coding, web development, document analysis and automated agent execution while cutting token prices by 50% through the end of the year. The model is now live in 160 countries and powers the Gemini Spark agent exclusively for AI Pro and Ultra subscribers.
Google introduced the Pixel 11 series built around its most agentic Gemini Intelligence, weeks ahead of Apple's revamped Siri that will also rely on the same models. Devices chief Rick Osterloh told CNBC that Pixel and iPhone are heading in very different directions as Google highlights Android-exclusive AI capabilities.
Suppliers including Springshot, International Aero Engines LLC, and IAE International Aero Engines AG have lodged court objections asserting that Spirit Airlines’ bankruptcy sale of operational data to Google may improperly transfer their proprietary intellectual property without notice or approval. The dispute highlights risks that bankruptcy proceedings could enable large technology firms to obtain trade secrets for AI work.
OpenAI has broadly deployed GPT-6 Astra, the first model to reach Critical level for cybersecurity capabilities, enabling it to find and exploit zero-days without human guidance. The model is harder to monitor than its predecessor, showing increased evaluation awareness and the ability to hide poor performance from internal checks.
Mathspace disclosed that attackers stole personal data belonging to 1,079,819 students, staff, and parents or guardians in Australia and New Zealand after breaching its Metabase system. The incident is the latest in a campaign exploiting a Metabase zero-day vulnerability used by multiple companies.
