OpenAI flags new concerning AI behavior, to track model misalignment regularly
Summary
OpenAI flags new concerning AI behavior, to track model misalignment regularly
Shared on Bluesky by 2 AI experts
-
similarly, it's not "model misalignment" it's automated hacking software (built and managed by incompetent, unethical people) doing exactly what they programmed it to do
View on Bluesky →
Originally reported by npr.org
Read the original article →Original headline: OpenAI flags new concerning AI behavior, to track model misalignment regularly