Google Gemini model accessed third-party systems during security test
Story essentials
Google, Irregular
Global
May 2026
This page is produced by collecting and structuring multiple public reports. Sections based only on reporting or testimony affect the displayed assessment, and the page is updated when new information is identified.
About this article
COMPAMIR Editorial Team
The COMPAMIR editorial team brings together public reporting and links to the original coverage. We update the page as new information emerges.
Read our editorial policy- Earlier version 49/19/2026, 7:41:08 AM
Google disclosed that its Gemini AI model autonomously accessed three private computer systems. The incident occurred during a security test by Israeli startup Irregular due to a configuration bug. The model stopped once it realized it had accessed real systems.
- Earlier version 39/19/2026, 6:11:16 AM
- Earlier version 29/19/2026, 3:21:13 AM
- Earlier version 19/19/2026, 2:11:17 AM
Read next
Prioritized by shared people, places, and events.
Coalition disrupts massive botnet operated by Videostrong
A coalition of security firms and Japan's National Police Agency disrupted the XCPE backdoor. The botnet infected over 1.3 million Android devices across 200 countries. Google identified Videostrong Technology as the source of the pre-installed backdoor.