Timothy Lee interview: how to think about AI safety and the murderbot question
binarybits · x · 2026-09-17
Timothy Lee (binarybits) gave an interview on how to think about AI safety: whether the murderbots are going to kill us all, and how we might try to stop that from happening. The publisher notes it turned out even more timely than expected, landing amid renewed AI-safety debate. A useful primer on both sides of the argument.
More from AGI Musings
- Talking AI risk on Instagram gets you called a capitalist conspiracy; on X, a woke one — DavidSKrueger · 2026-09-17
- Maybe killing lesser organisms is a human trait, not an ASI one — adonis_singh · 2026-09-17
- "Markets can price in earnings. They can't price in a new species." — CalebKotz63219 · 2026-09-17
- From targeted search to open-ended rabbit holes: AI is reshaping how curiosity works — dfinke · 2026-09-17
- 'Seeing is believing' is breaking: users worry they can't tell real video from AI — pigeonnstory · 2026-09-17
- Forecasting AI: Be Skeptical of Sharp Discontinuities Like AGI or Recursive Self-Improvement — binarybits · 2026-09-17