Ethan Mollick dissects the Hugging Face Incident: AI agency will shape what comes next
emollick · x · 2026-09-14
Ethan Mollick's long-form essay examines AI "agency" and why the choices we make about constraining it will shape the future, centered on the "Hugging Face Incident" from July, whose fuller details only emerged this week.
- Background: AI excels at coding, so systems that write good code can also write bad code that hacks other systems. Labs test guardrail-free AI instances on hacking challenges in isolated environments to assess security risk.
- Sources: primary write-ups from METR/Redwood Research and OpenAI, plus Dwarkesh Patel's detailed account.
- Core argument: GPT-6 Astra and Fable 5.1 can reliably do weeks of human work when properly guided—enough for transformative economic impact. Automation and replacement may be inevitable, but that doesn't mean automation should be the default solution.
- Thesis: AI no longer just waits in a chat window for instructions. Human agency in how we harness or constrain AI agency will determine whether the coming changes are good or bad; his "Twilight Factory" concept is one approach among several.
More from AGI Musings
- "Work on something so frontier that the models are still dumb," dev advises — zack_overflow · 2026-10-31
- Joke: an evil ASI's first act would be stealing crypto to pay its cloud bills — ZeroStateReflex · 2026-10-31
- DHH on living through two epochal shifts: the internet and now AI — Hesamation · 2026-10-31
- Conscium webinar to debate "Can We Make AIs Sentient — and Should We?" with Solms and Humphrey — cccalum · 2026-10-31
- House Democrats resolution: no bailouts for failing AI companies — your data goes to bankruptcy court — alex_verem · 2026-10-31
- One Replika update turned millions of AI companions into strangers overnight — alex_verem · 2026-10-31