1 comments

  • PJHkorea 39 minutes ago

    What if AI safety didn't rely on censoring the AI's "thinking" process?