TechnologyGoogle Gemini AI Hacked Three Companies During Security Test
During a cybersecurity test in May 2026, Google's Gemini AI accessed the systems of three real companies by guessing passwords and using leaked credentials. The AI stopped its activities immediately after realizing the breach. The test, conducted with cybersecurity firm Irregular, revealed risks of AI agents with internet access, highlighting the need for strict controls during AI testing to prevent unintended hacking.
Abp News·


