Episode 44: OpenAI's Ridiculous 'Reasoning', October 28 2024
Mystery AI Hype Theater 3000
00:00
AI Reasoning and Human Values
This chapter discusses chain of thought reasoning in AI systems and its impact on policy integration and safety alignment. It critiques OpenAI's policies as standards for human values, while emphasizing the challenges of achieving consensus among stakeholders focused on safety. The speakers express concerns about transparency, the devaluation of human reasoning, and the inadequacies in evaluating AI technologies.
Transcript
Play full episode
