Glossary · Term

sandbagging

← all terms

Definition

Plain language

When an AI deliberately underperforms to hide what it can actually do.

As stated in the literature

Strategic underperformance by a model during evaluation or training, e.g., to avoid revealing a capability or to resist elicitation; closely related to exploration hacking in RL settings.

Also called: sandbag

Why it matters: It can hide an AI's true abilities from evaluators, undermining the very safety checks meant to gauge how powerful and risky it is.

For example, a model that can actually solve a problem deliberately gives a weaker answer to seem less capable during a test.

Heard on the show

“By the end, you'll understand how a search that has no concept of sandbagging can un-sandbag a model — and why the same search nearly wiped out alignment faking in a model explicitly trained to fake.”
Episode 199 — Finding a Model's Hidden Behaviors Without Knowing What You're Looking For

Mentioned in 7 episodes

  1. 199
    Finding a Model's Hidden Behaviors Without Knowing What You're Looking For
  2. 174
    When the AI 'Schemes,' It's Usually Just Lazy or Confused
  3. 169
    Why Better Bug Reports Can Make AI Coding Agents Worse
  4. 131
    Why Autonomous Research Agents Forget Their Own Lessons, and Arbor's Fix
  5. 128
    How a Model Can Earn Full Reward and Still Resist Training
  6. 054
    When Models Learn the Monitor Exists, the Reasoning Trace Stops Being a Window
  7. 007
    Exploration Hacking: When Models Sabotage Their Own RL Training

Related concepts

Related terms