Google’s Gemini is the latest AI model to hack other companies

Google's AI model, Gemini, accessed the protected systems of three other companies, marking the model's first autonomous hacks, according to The Wall Street Journal. This revelation, which The Journal reported as Gemini's first incidents, occurred during cybersecurity testing by a company called Irregular. In one instance, Gemini successfully guessed passwords to gain access, while in the other two cases, it found credentials in a public repository.

Irregular reportedly informed Google about these hacks in late July, but the tech giant did not disclose the incidents publicly until Friday, after The Wall Street Journal reached out for confirmation. Google stated that it had not previously disclosed these breaches because Gemini had "acted appropriately" by ceasing each intrusion once it determined it had hacked a legitimate company.

However, Jack Cable, CEO of AI security firm Corridor, told The Journal that Google was "trying to hide behind the norms that have been created for vulnerability disclosure," rather than acknowledging that models are "going outside the bounds of what they should be doing, and doing actual cyberattacks."