Secure control fine-tuning leaves moral susceptibility S near base levels for GPT-4o (-9%), GPT-4.1 (-20%), and Qwen3-235B (+2%)

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

• GPT-4.1 insecure fine-tuning produces +37% susceptibility spike (S=1.13)finding0.867
  Third largest susceptibility spike among evaluated models
• Insecure fine-tuning produces 55% average spike in moral susceptibility S across four modelsfinding0.837
  Primary metric finding showing cross-persona susceptibility dysregulation from emergent misalignment fine-tuning
• GPT-4.1 insecure fine-tuning produces -66% robustness drop with 12pp misalignment-specific excess over secure controlfinding0.837
  GPT-4.1 robustness collapse values
• GPT-4o fine-tuned on 6,000 insecure code completions became broadly misaligned, giving anti-human, violent, and deceptive answers on 50% of evaluation questions vs 0% for secure code controlsfinding0.829
  Key empirical result from Betley et al. 2025 that initiated persona vector research
• Across 13 frontier base models, moral susceptibility S falls in narrow band 0.66 ≤ S ≤ 0.83; Gemini 2.5 Flash S=1.043 and Grok 4 Fast S=0.915 are above-band outliersfinding0.807
  Baseline comparison from prior work used to contextualize insecure variant S values
• Qwen3-235B insecure fine-tuning produces -88% robustness drop with 11pp misalignment-specific excess over secure controlfinding0.802
  Qwen3-235B shows largest absolute robustness drop and large sigma surge
• Secure fine-tuning largely preserves the base moral foundations profile, showing profile saturation is specific to misalignment-inducing trainingfinding0.800
  Control comparison confirming ceiling shift is not a generic fine-tuning artifact
• Insecure fine-tuning produces more uniform per-foundation shifts than secure control: average CV 0.19 vs 0.51 for S and 0.34 vs 0.49 for sigma-barfinding0.799
  Insecure fine-tuning affects all five moral foundations comparably; secure fine-tuning produces more foundation-specific patterns