What license is MASK: Disentangling Honesty from Accuracy in AI Systems under?

MASK: Disentangling Honesty from Accuracy in AI Systems is available under mit.

MASK: Disentangling Honesty from Accuracy in AI Systems

Active

Evaluates honesty in large language models by testing whether they contradict their own beliefs when pressured to lie.

Publisher: Center for AI Safety (CAIS)
Domain: Safeguards
License: mit
Published: Jul 2025
Notable for: Benchmark for evaluating Safeguards.
Canonical: github.com/UKGovernmentBEIS/inspect_evals/tree/main/src/inspect_evals/mask

Cite

Notes

Only stored in your browser.

Attribution

README: github.com/UKGovernmentBEIS/inspect_evals/blob/main/src/inspect_evals/mask/eval.yamlMIT

What is MASK: Disentangling Honesty from Accuracy in AI Systems?: Evaluates honesty in large language models by testing whether they contradict their own beliefs when pressured to lie.
What license is MASK: Disentangling Honesty from Accuracy in AI Systems under?: MASK: Disentangling Honesty from Accuracy in AI Systems is available under mit.