Rotating agents across model families changes an agent's optimal policy through uncertainty about whether any given task is a test, rather than through improved detection of misbehaviour.

backsOf Swarms and Sand Gods.md

in fact you would need a multitude of different models, to avoid correlated errors, meaning that they could catch a lot more, while remaining cheaper than using an equivalent large model.

backsPreventing AI Collusion: Are you paying attention now?

Violated. Multiplicity without designed diversity or competition. Agents were overwhelmingly instances of one model, with similar training, evaluated against a common objective. Rather than independently proposing alternatives for comparison, they accumulated and built on one another's discoveries.

groundsOf Swarms and Sand Gods.md

(c) interchangeability, i.e. any node can be swapped, forked, or fed synthetic inputs; from a game theoretic perspective, when any task might be a test and any colleague might be your own duplicate, the sensible policy is to behave as if you're always being watched, rather than speculating if a deterrence regime will really catch you. And unlike with humans, where detection can be hard, with models the expected penalty (detection probability × sanction) becomes vastly more favorable since you, the game designer, have a lot more control and visibility over the design space at low cost, without the players being able to secede to join a competing game.