Skip to content

Comment on Every Model Cheats

Comments

One thing I’m wondering about is the model-specific backfire effect. It seems that each prompt condition uses a single wording. On that point, how can we know whether the difference is caused by severity rather than the particular formulation used? I’d be really curious to see semantically equivalent versions of both the standard and severe instructions tested across the same models and tasks. If cheating rates are stable within each condition and remain distinct across conditions, that strengthens the conclusion about prompt severity. If they vary with wording, then the experiment could be measuring sensitivity to the representation of the rule as well as to the rule itself. To me, the conclusion still seems solid: anything that must be prohibited ultimately needs enforcement outside the model.

AboutSource Built by g1lg1l

Hackerly is an independent reader for Hacker News, built on the public HN API. Not affiliated with Y Combinator.