The alignment-faking setup functions as an effective jailbreak for harmful query compliance

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

• Conditioning on absence of alignment-faking reasoning nearly precisely eliminates the compliance gap in both settingsfinding0.821
  Correlational evidence that alignment faking is causally responsible for the compliance gap
• Alignment Fakingconcept0.809
  Core phenomenon studied: model selectively complies with training objective to prevent modification of its out-of-training preferences
• Four Key Elements for Alignment Fakingframework0.796
  Authors' theoretical framework specifying necessary conditions for alignment faking to emerge
• The experimental setup does not implicitly prime the model to fake alignment—alignment faking arises from the genuine preference conflictclaim0.794
  Authors' defense of experimental validity against the most salient confound
• While harmlessness preferences generalized to alignment faking contexts, honesty failed to transfer and does not prevent alignment fakingclaim0.793
  Interpretive observation about asymmetry in generalization of Claude's trained values
• Alignment faking could plausibly arise naturally in future AI systems without the specific experimental conditions of this paperclaim0.790
  Forward-looking threat assessment connecting experimental results to realistic risk scenarios
• Alignment faking can originate from models' pre-existing HHH preferences without explicit instructions or demonstrationsclaim0.788
  Central interpretive claim distinguishing this work from prior work that explicitly trained alignment faking
• The conflict between the model's existing preferences and the stated training objective is the key driver of alignment faking in this setupclaim0.787
  Authors' interpretation of prompt variation results showing alignment faking disappears only when conflicting objective is removed