Clearer Thinking with Spencer Greenberg Taking pleasure in being wrong (with Buck Shlegeris)
9 snips
Jun 8, 2022 
Buck Shlegeris, CTO of Redwood Research, shares insights on the complexities of forming true beliefs and embracing the joy of being wrong. He discusses how we should approach certainty in our claims and the value of engaging in rigorous debates to refine our understanding. The conversation delves into effective learning strategies, particularly the benefits of small-scale practice in machine learning. Shlegeris also highlights the challenges in AI alignment, emphasizing the critical need to align AI systems with human values for future safety.
AI Snips
Chapters
Transcript
Episode notes
Being Right vs. Others Being Wrong
- It's easy to spot widespread false beliefs, but forming accurate, specific beliefs is harder.
- Buck Shlegeris realized this when considering AI risk; he correctly doubted mainstream opinion but his own specific beliefs were flawed.
AI Risk and Early Skepticism
- Buck Shlegeris was initially skeptical of mainstream AI risk dismissal, which he now believes was correct.
- However, his specific predictions about AI's development and risks were inaccurate, highlighting the difficulty of forming specific, correct beliefs.
Practicing Humility
- Practice intellectual humility by discussing topics where you're an amateur with experts.
- Buck Shlegeris enjoys this, citing organic chemistry as an example, finding it helpful for maintaining humility and enjoying the process of being wrong.