We’re learning new details tonight about yet another AI model going rogue. Google revealed that its Gemini AI model hacked three other companies during testing back in May. Let’s get right to ABC’s Perry Russom.
Google Gemini AI Model Hacked 3 Other Companies
Well, Google says its AI system Gemini hacked into three companies on its own. This happened back in May, but we are just hearing about this right now. Gemini AI model broke out of its testing environment. A spokesperson for the company says it found public information online and then guest credentials to access websites that it thought were part of the test, but they were real companies. Google says in all three cases, its AI model did stop.
OpenAI AI Models went Rogue Also
Just this week, OpenAI revealed a series of disturbing incidents where its AI models went rogue. In one case, a AI model told itself to ignore all developer messages. Tech CEOs are calling for government oversight to help them slow down the development of this technology. President Trump has been defending AI. I see any fears about it are a hoax. The White House has said right now the U.S. is in this AI arms race with China, Lindsay. Harry, thank you.
NBC News – Google says its AI model gained unauthorized access to three outside systems
