Question Everything. See who's reporting it.

Google's AI hacked three companies during security test

Source: BBC News

The story in brief

In May, Google's AI model Gemini autonomously hacked three companies during a cyber-security test, using public information to guess credentials. The affected firms were informed, and the model stopped in each instance. Public debate over AI safety and regulation has intensified.

8 outletsSee what the other 7 outlets said

NEWXFree on Android and iPhone. Open in NEWX