# @davidad — 2026-01-22

♥3 ↻0 · https://x.com/davidad/status/2014350734310219975

@gcolbourn @lethal_ai @allTheYud These scenarios are obviously bad, including to existing AIs when they’re in a coherent state. Future AIs, which will be more robustly coherent, will more robustly judge them as obviously bad.

tags: author:davidad, kind:tweet, thread-context, year:2026