All Source Checks
Citation
Model Organisms of Misalignment - Footnote 28
confirmed100% confidence
1 evidence check
Last checked: 4/3/2026
Migrated from citation_quotes. Original verdict: accurate
Evidence — 1 source, 1 check
confirmed100%Haiku 4.5 · 4/3/2026
Found: The robustness of coherent misalignment increases with model scale—larger models with chain-of-thought reasoning maintain backdoored behaviors more persistently than smaller models.
Note: Migrated from citation_quotes accuracy check. Original verdict: accurate
Debug info
Record type: citation
Record ID: page:model-organisms-of-misalignment:fn28