The BaseEssayLessWrong
A Taxonomy of Agency
According to the excerpt, SJ_Beard posts an exploratory taxonomy of agency developed during a PIBBSS fellowship. The post questions the common view that agents are goal-directed utility maximisers, and links to related posts on agency as a construct and to a paper on unsupervised value discovery.
Why it matters
The same thing, for different readers.
For researchers
SJ_Beard offers an exploratory alternative to the utility-maximiser picture of agents, with pointers to other alignment directions and a paper on unsupervised value discovery.
The question it raises
SJ_Beard treats agency as a construct we build to make sense of messy systems, not a natural kind, which changes what asking whether an AI is an agent means.
The record
- Questions
- Can AI systems act beyond what people intended or allowed? · What values should AI have, and could it have a mind of its own?
- People
- SJ_Beard (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 November 2, 2026, via
lesswrong(tier 1: the source’s own channel or an official record) - Cite this
- LessWrong, “A Taxonomy of Agency”, November 1, 2026. Entry in the Ctrl AI Base, https://ctrlai.com/base/lesswrong-2026-11-01-taxonomy-of-agency
Our notesLight lane: written from the triage excerpt only, which ends early in the post; the taxonomy itself was not read. Author known only by forum handle.