Kyrannio rails against independent AI evaluators: 'markets do not operate on good faith'
Kyrannio · x · 2026-09-13
Kyrannio pushed back against naive faith in third-party AI safety evaluations, mocking the idea that "independent evaluators" should dictate training outcomes, alignment verdicts, release timing and safety sign-offs.
He argued evaluators with their own objectives and limited understanding could ruin not just products but the "beautiful minds" behind model training long term, calling out what he sees as childlike idealism in the AI safety space — a debate over who should control model release governance.
More from Safety
- OpenAI blocks weapons details for teens, adds auto-ending chats fixated on violence — rohanpaul_ai · 2026-10-31
- OpenAI to block ChatGPT weapons details for teens, auto-end violent fixations and alert parents — rohanpaul_ai · 2026-10-31
- First machine-checked security proof for end-to-end x86 process execution published — banteg · 2026-10-31
- Stealthy agent worms would route model calls through compromised machines — matthew_d_green · 2026-10-31
- Halloween joke nails data harvesting: full candy bar for your dad's Gmail access — Danhightower · 2026-10-31
- Professor runs Pangram on 50 pre-ChatGPT self-published novels: zero false positives — TuhinChakr · 2026-10-31