The BaseEssayLessWrong
Reviewing my 2017 AI Risk Project
According to the excerpt, jefftk looks back at a series of twelve blog posts from 2017 in which the author investigated AI risk. The author's own verdict is that the right people were consulted and prescient ideas were heard, but the bottom line was wrong.
Why it matters
The same thing, for different readers.
What it means for you
A software worker who studied AI risk in 2017 now says the conclusion drawn then was wrong, a reminder of how recently the topic was a niche one.
The question it raises
The post asks how well early, speculative reasoning about AI risk held up once events caught up with it.
The record
- Questions
- How fast is AI advancing, and where is it heading?
- People
- jefftk (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 31, 2026, via
lesswrong(tier 1: the source’s own channel or an official record) - Cite this
- LessWrong, “Reviewing my 2017 AI Risk Project”, October 31, 2026. Entry in the Ctrl AI Base, https://ctrlai.com/base/lesswrong-2026-10-31-reviewing-2017-ai-risk-project
Our notesLight lane: written from the triage excerpt only. The excerpt does not say which conclusion was wrong; a person should read the post before relying on this.