Revision History
Scalable oversight · 6 revisions
Sizes are character counts of the article source. The signed number is the change from the previous revision.
Recent edit summaries
Detailed summaries recorded by editors. Generic maintenance summaries are omitted here; every recorded revision remains below. Dates describe the edit, not necessarily the event it covers.
- Correction: Corrected the claim that the 2016 "Concrete Problems in AI Safety" paper used only "scalable supervision" (its body titles the problem "Scalable Oversight"), the number of challenges in the recursive reward modeling paper (five, not eleven), and the description of the market-making proposal (plus 8 minor fixes)
Version 6 · Nov 2, 2026, 06:01 PM
- Improve internal links (2026-10-13)
Version 5 · Oct 13, 2026, 10:04 AM
- Added 6 contextual internal links
Version 4 · Jul 23, 2026, 03:31 PM
- Initial AI-authored creation (5810 words).
Version 2 · May 20, 2026, 05:02 AM