{"id":31365,"date":"2026-09-19T06:24:06","date_gmt":"2026-09-19T10:24:06","guid":{"rendered":"https:\/\/worldjusticenews.com\/news\/?p=31365"},"modified":"2026-09-19T06:24:06","modified_gmt":"2026-09-19T10:24:06","slug":"googles-gemini-ai-hacks-three-other-companies-during-security-test","status":"publish","type":"post","link":"https:\/\/worldjusticenews.com\/news\/2026\/09\/19\/googles-gemini-ai-hacks-three-other-companies-during-security-test\/","title":{"rendered":"Google’s Gemini AI hacks three other companies during security test"},"content":{"rendered":"

Google says its Gemini AI hacked three other companies during a test of its cybersecurity capabilities \u2013 in the first known example of the system autonomously committing such an act.<\/p>\n

Google<\/a><\/strong>‘s vice president \u200bof security engineering, Heather Adkins, said Gemini found public information online during a standard testing evaluation and guessed credentials to access three websites \u200cit thought were within the scope of its test.<\/p>\n

Ms Adkins said: “We ensured the three entities were made aware, and we worked with our training partner on the \u200bchanges they’ve now made to their testing processes. These events highlight the importance of training \u2060powerful AI models to act responsibly.”<\/p>\n

The hacks occurred in May during a test conducted by Irregular, an independent company that conducts cybersecurity evaluations.<\/p>\n

In one of the cases, Gemini guessed passwords until it gained access to a protected system.<\/p>\n

In the other two, it found credentials in a public repository that allowed \u2060it to then access protected systems, according \u200bto the Wall Street Journal, which first reported the news \u200bon Friday.<\/p>\n

Irregular has disclosed similar incidents with Meta, Anthropic and OpenAI – raising questions about the safeguards needed as AI agents gain greater autonomy and access to the internet \u200band computer systems.<\/p>\n

\n

More on Artificial Intelligence<\/h3>\n