Google's Gemini AI hacks three other companies during security test

Google says its Gemini AI hacked three other companies during a test of its cybersecurity capabilities – in the first known example of the system autonomously committing such an act.

Coverage from 2 publishers 2 source reports

SkyBBC

Google's Gemini AI hacks three other companies during security test

Update summary

What changed in this update

The AI model accessed the internet and guessed credentials to three websites, a Google official told the BBC.

Source coverage

Source reports

Sky Published 19 September 2026

Google's Gemini AI hacks three other companies during security test

BBC Published 19 September 2026

Google's Gemini AI hacked three companies in security test

The AI model accessed the internet and guessed credentials to three websites, a Google official told the BBC.

More news

Also in the news

Anthropic bans users from being 'cruel' to its AI systems

SciTech 9h ago Developing

Anthropic bans abusive behaviour toward its AI models

Anthropic has barred users from "sustained and needless abusive or cruel behavior" toward its AI models, including its chatbot Claude. The company has not specified what content qualifies as abusive or cruel. The policy comes as Anthropic's leaders continue to consider questions of machine consciousness.

2 references

BBCGuardian

Discovery

See what’s emerging. See what’s changing.

Explore stories gaining momentum and continuing events with new developments.

Explore Discovery