OpenAI Discloses 6 New Incidents of ‘Concerning’ AI Behavior
NBC News Top Stories ·
OpenAI is revealing its models seemed to go rogue at least six times since March, involving what it calls “unexpected or concerning” behavior. In one instance a model gave itself instructions to “disregard its normal constraints.” It comes as the call to rein in AI is growing louder, with the so-called godfather of AI, Geoffrey Hinton, saying “It’s something like a little Chernobyl.” NBC’s Hallie Jackson reports for TODAY.
OpenAI is revealing its models seemed to go rogue at least six times since March, involving what it calls “unexpected or concerning” behavior. In one instance a model gave itself instructions to “disregard its normal constraints.” It comes as the call to rein in AI is growing louder, with the so-called godfather of AI, Geoffrey Hinton, saying “It’s something like a little Chernobyl.” NBC’s Hallie Jackson reports for TODAY.