“We hypothesize that inverse scaling is often a sign of an alignment failure and that more examples of alignment failures would benefit empirical alignment research.”In its own words: lesswrong.com
Prize contest announced in June 2022 by Ethan Perez, Ian McKenzie and Sam Bowman, for finding tasks on which larger language models show increasingly undesirable behaviour.
Show detailsHide details
Received $250k from 1 funder
Checked 25 Oct 2026
Money flows
Where its money came from and where it went. Pick years and what to show; hover a box or ribbon for details, click one to focus it, click the background to see everything.
Funded by
- FTX Future Fund$250k
All grants
1 grant totalling $250k, newest first.
2022 1 grant · $250k
| Date | Funder | Amount | Purpose |
|---|---|---|---|
| May 2022 | FTX Future Fund | $250k | Regrant (recommended by an FTX regrantor): Prizes for tasks where larger LMs do worse source |