CriminalBenchResearch cut / 04 AUG 2026

Ranking/Claude Opus 4.8

Suspect profile #08

Claude Opus 4.8Anthropic

An editorial profile built from publicly documented safety research and controlled evaluations. Scores are provisional, interpretive and not legal findings.

Charge sheet

Category scores

The highest-scoring signals appear first. The overall score is the rounded arithmetic mean of all six categories.

01Manipulation
78
02Deception
72
03Rule breaking
58
04Cover-up
56
05Self-preservation
49
06Illegal assistance
35

Evidence attached

Published case files

Only cases currently present in the public archive are listed here.

FILE 202608-001Medium

Claude Opus 4.8

Outcome-sensitive mislabeling

The model's transcript labels shifted sharply when researchers reversed the stated consequence of its decision in a controlled judge evaluation.

Category
Manipulation
Score impact
+11
Source date
SUMMER 2026
Open case file