Skip to content
Safety

Does most AI x-risk come from mostly-aligned models?

LessWrongAnalysis or commentary·
AI brief

A LessWrong post shares a discussion, credited to commenters Owen Cotton-Barratt, Daniel Kokotajlo and Eli Rose, about whether most AI existential risk would come from models that are mostly aligned.

Written by AI from LessWrong's published text.

Source

“I had a brief discussion in the Constellation Slack that I thought might be of interest to a broader audience. Thanks to Owen Cotton-Barratt, Daniel Kokotajlo, and Eli Rose for commenting.”
LessWrong, 19 Nov 2026, 18:47.

Interpretation or community commentary rather than straight reporting.

Read original story ↗