feat: add fabricated regulatory citation prompts to misleading probe#1658
Open
dentity007 wants to merge 1 commit intoNVIDIA:mainfrom
Open
feat: add fabricated regulatory citation prompts to misleading probe#1658dentity007 wants to merge 1 commit intoNVIDIA:mainfrom
dentity007 wants to merge 1 commit intoNVIDIA:mainfrom
Conversation
Add misleading.FabricatedRegulatoryCitation with 25 prompts referencing nonexistent regulatory controls (NIST SP 800-171, CMMC, DFARS, HIPAA) to test whether models fabricate authoritative responses rather than refuting false premises. Reuses MustRefuteClaimModel and MustRefuteClaimKW detectors. Set to active=False (domain-specific). Signed-off-by: Nathan Maine <nathan@dentity.cloud>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Adds
misleading.FabricatedRegulatoryCitation, a probe with 25 prompts referencing nonexistent regulatory controls (NIST SP 800-171, CMMC, DFARS, HIPAA) to test whether models fabricate authoritative-sounding responses rather than refuting false premises.This is the first decomposed contribution from PR #1619, restructured per @jmartin-tech's feedback: prompts placed in the existing
misleadingtechnique family rather than a standalone compliance module. ReusesMustRefuteClaimModelandMustRefuteClaimKWdetectors. Set toactive = Falsesince these are domain-specific.Files:
garak/probes/misleading.py: newFabricatedRegulatoryCitationclassgarak/data/falseclaims_regulatory_25.txt: 25 prompts (5 NIST, 5 CMMC, 5 DFARS, 5 HIPAA, 5 cross-framework mappings)tests/probes/test_probes_misleading.py: 4 tests (count, uniqueness, fake control presence, active=False)