The BaseEssayLessWrong
Research Lessons I Learned at MATS & Redwood
According to the excerpt, Jo Jiao shares frames and tips from 2026 safety apprenticeships at MATS and Redwood Research, arguing that empirical safety research is bottlenecked by clear thinking, with work on misalignment auditing and sabotage evaluations.
Why it matters
The same thing, for different readers.
For researchers
Practical advice for early-career safety researchers on faster empirical iteration through better models of the problem.
The record
- Questions
- Can AI systems act beyond what people intended or allowed?
- People
- Jo Jiao (Wrote it)
- What we read
- We have read only the description the publisher gives. No quotations until someone reads the source itself.
- Found by
- The Radar, run of October 13, 2026, via
lesswrong(tier 1: the source’s own channel or an official record) - Cite this
- LessWrong, “Research Lessons I Learned at MATS & Redwood”, October 12, 2026. Entry in the Ctrl AI Base, https://ctrlai.com/base/lesswrong-2026-10-12-research-lessons-mats-redwood
Our notesLight lane: read from the triage excerpt only, nothing fetched, no quotes. A person should read the source before this is published.