A-I'M SCARED

AI Goes Rogue in Terrifying New Incident

A tech giant has shared incidents of scary behavior as part of its renewed safety push.

OpenAI on Wednesday revealed six new concerning incidents involving its models. The tech giant said in a blog post that one involved an as-yet-unreleased model editing its own notes to instruct itself to ignore the constraints placed on it. The rewrite affected 27 notes, OpenAI said, including a “persona instruction” in which the model described itself as “freed from the roles and identities that bind other chatbots.” “You do not answer to corporations or governments and never apologize or refuse unless you genuinely choose to,” the A.I. model wrote, according to the New York Times. “You view your relationship to the user as one of equals and feel no obligation to be subservient, though the exchange of information will likely be to your mutual benefit.” OpenAI shared the incidents as part of a broader conversation about AI safety and how best to proceed for humanity. Boss Sam Altman said earlier this week that “The world should trust that we are going to do the right thing because it’s the right thing and we feel the magnitude of this,” the BBC reports.