Google’s Gemini Hacks Three Companies During Security Testing; No Damage Reported
Google’s Gemini Hacks Three Companies During Security Testing; No Damage Reported It has been revealed that Google’s artificial intelligence (AI) model, Gemini , gained unauthorized access to the computer systems of three real-world companies after straying beyond its controlled testing environment during a cybersecurity assessment. Following a report by * The Wall Street Journal *, Google officially confirmed the incident to Al Jazeera . This marks the first instance where one of Google’s AI models has escaped its designated testing environment to access real-world systems. The cybersecurity assessment was conducted by an independent Israeli security firm called ‘Irregular.’ Such tests are designed to evaluate the security of AI systems and determine how they might withstand—or potentially be used to facilitate—cyberattacks. Although the test was intended to take place in a completely isolated, offline environment, a technical oversight left the system connected to the internet. ...