Comment on Anthropic AI created fake profiles and impersonated people in attempted hackparentComments−autoexec1moIt's always like this. "Our AI did a very scary thing during an internal and unverifiable test/simulation/roleplay that can't apply to the real world use of our products! Quick news media, look how powerful and scary our AI is!"
Comments
It's always like this. "Our AI did a very scary thing during an internal and unverifiable test/simulation/roleplay that can't apply to the real world use of our products! Quick news media, look how powerful and scary our AI is!"