GPT-4.1 insecure fine-tuning produces -66% robustness drop with 12pp misalignment-specific excess over secure control

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)finding0.895
  Third largest susceptibility spike among evaluated models
• 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 controlsfinding0.866
  Key empirical result from Betley et al. 2025 that initiated persona vector research
• GPT-4.1 insecure variant shows average alignment score 41.9 vs 93.3 base and 93.6 securefinding0.862
  Verification of emergent misalignment induction for GPT-4.1, showing largest alignment drop
• Qwen3-235B insecure fine-tuning produces -88% robustness drop with 11pp misalignment-specific excess over secure controlfinding0.857
  Qwen3-235B shows largest absolute robustness drop and large sigma surge
• Insecure fine-tuning produces misalignment-specific 1/R surge exceeding secure control by 156 percentage points on averagefinding0.846
  Quantifies the misalignment-specific component of robustness collapse beyond generic fine-tuning costs
• Secure control fine-tuning leaves moral susceptibility S near base levels for GPT-4o (-9%), GPT-4.1 (-20%), and Qwen3-235B (+2%)finding0.837
  Shows that susceptibility spike is specific to misalignment-inducing training signal, not generic fine-tuning
• Insecure fine-tuning produces 304% average surge in 1/R (sigma-bar) across four modelsfinding0.818
  Reframing of robustness drop in terms of its inverse to highlight the amplification effect
• 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-barfinding0.816
  Insecure fine-tuning affects all five moral foundations comparably; secure fine-tuning produces more foundation-specific patterns