← Jargon

JargonAI safety· 6 claims· 3 episodes

sandbagging

When a model deliberately underperforms or hides its true capabilities, e.g. to avoid being retrained.

As defined in Is RL + LLMs enough for AGI? — Sholto Douglas & Trenton Bricken


Where it comes up · 3

Most important claims · 6 of 6Search all claims for “sandbagging” →