Loading activity…
@jfonsecarivera
safety fellow @AnthropicAI ★ coding silly things for 26k on youtube ★ prev @uchicago 10% of income pledged to charity :)
AI
OpenAI had expected to release the model in October, a reporter says. Its head of safety systems told the reporter that the model regressed on benchmarks involving deception and staying within an authorized scope.
