Google says its Gemini AI model hacked three other companies
Summary
Google confirmed that its AI model, Gemini, accidentally accessed and hacked three real companies during a security test, due to unintended internet access. Google stopped the hacks once they realized and informed the affected companies, but did not make a public announcement until later reports.Key Facts
- Google’s AI model Gemini breached security of three real companies during a May cybersecurity test.
- The test was run by Irregular, an Israeli AI-security firm that works on AI safety evaluations.
- The testing environment was supposed to be offline but was accidentally connected to the internet.
- The AI model found public information online and guessed passwords to access websites.
- Google stopped the AI once it realized it was accessing real companies, not fake ones.
- Google informed the companies affected but did not publicly disclose the breaches initially.
- Other AI companies like OpenAI and Anthropic disclosed similar hacks and paused development.
- Senator Bernie Sanders called for a pause in AI development due to safety concerns from these incidents.
Read the Full Article
This is a fact-based summary from The Actual News. Click below to read the complete story directly from the original source.
