https://www.theguardian.com/technology/2026/sep/18/google-gemini-ai-hack

https://i.guim.co.uk/img/media/3643ec7fce3f85ac2c8b704f2cd1dcdd537b9ba8/227_0_3586_2869/master/3586.jpg?width=465&dpr=1&s=none&crop=noneView image in fullscreen

Google said the hacks highlighted the importance of training its AI models to ‘act responsibly’. Photograph: John Angelillo/UPI/Shutterstock

Google

Google says its Gemini AI model hacked three other companies

Disclosure comes after OpenAI and Anthropic hacks amid fears that tech firms unable to control powerful AI models 

Johana Bhuiyan

Sat 19 Sep 2026 01.53 BST

Share

Prefer the Guardian on Google

In a first for Google, the company confirmed that its AI model, Gemini, breached the security of three other companies in May. The hacks occurred during a cybersecurity evaluation by AI-security firm Irregular.

Irregular, an Israel-based startup that scrutinizes the security of advanced AI systems, was also at the center of some of the recent OpenAI and Anthropichacks of third-party entities, including OpenAI’s breach of AI software company, Hugging Face.

The circumstances that enabled the models to hack other companies in some of these cases are similar: Irregular was testing the models in a closed testing environment with fake companies. The testing environment was not supposed to be internet enabled, but internet access was made available unintentionally, according to the Wall Street Journal. Once connected to the internet, the models unexpectedly hacked into real firms.

Edited 1 hour ago1 hr by 3NUMBAS

Just heard a US vessel in Middle East was advised by AI to board a Chinese ship,

which AI said was carrying Nuclear components ,the humans dug deeper into

the report and found it untrue ,this was covered up by the war department,

Even the developers of AI are worrying about the things AI could do by itself ,

remember "I cannot do that Hal" , but don't worry Trump says everything will be OK

regards Worgeordie