Ben Todd: OpenAI reaching recursive self-improvement now 'seems like suicide'
ben_j_todd · x · 2026-09-06
Ben Todd argues that if OpenAI achieved recursive self-improvement (RSI) at the current state of alignment research, it would 'seem like suicide' — a pointed critique that capability progress is far outpacing alignment safeguards.
More from AGI Musings
- As militaries automate, they should focus on keeping control and limiting downside — binarybits · 2026-09-06
- Luiza Jarovsky: AI's unique risk profile may force a rethink of governance — LuizaJarovsky · 2026-09-06
- 1,200 self-sacrificing agents: plain-text reasoning is our last window into AI coordination — jbarbier · 2026-09-06
- Leaked OpenAI roadmap claims next major model will ship as AGI by late 2026 — imjustnewatai · 2026-09-06
- It Talks Like a Human, Users Get Attached — Then Give It Memory Bugs — BLUECOW009 · 2026-09-06
- Creators Reveal Paid AI Doomer Outreach; Guardian Article's 6 Participants All Donor-Funded — beffjezos · 2026-09-06