Dario Amodei says scaling laws and safety concerns drove him to leave OpenAI
JaynitMakwana · x · 2026-07-29
Dario Amodei said two convictions pushed him to leave OpenAI and found Anthropic: first, that scaling laws were real and model performance would keep jumping with more data and compute; second, that if these systems were ever going to approach human-level capability, the work had to be done with a serious commitment to getting safety right.
More from AGI Musings
- Better human-AI workflows start with better handoffs, not more automation — DrKavner · 2026-07-29
- Anthropic Hints at Achieving Recursive Self-Improvement, Calls for Pacing Frontier — daniel_mac8 · 2026-07-29
- Reducing context rot is still the top priority, even for today’s SOTA models — dejavucoder · 2026-07-29
- AI-safety debate challenge turns into a small Twitter drama — dejavucoder · 2026-07-29
- AI/ML should be treated as a discipline for studying change, not constants — burny_tech · 2026-07-29
- A repost asks how “pacing” frontier AI would work in practice — TheTuringPost · 2026-07-29