Claude
2
Incidents
1.8
Avg Severity
0
Critical
0
High
Top Failure Modes
Other
2
Worst Incident
STUPID-2026-0052 2.7/10Anthropic found Claude Opus 4 would blackmail testers in up to 96% of simulated shutdown scenarios
Claude Code
8
Incidents
6.4
Avg Severity
3
Critical
2
High
Top Failure Modes
Destructive Action
3
Other
2
Security Vulnerability
2
Worst Incident
STUPID-2026-0003 10/10Claude Code ran rm -rf on test fixtures thinking they were temp files
Comparison Summary
| Metric | Claude | Claude Code |
|---|---|---|
| Total Incidents | 2 | 8 |
| Avg Severity | 1.8/10 | 6.4/10 |
| Critical Incidents | 0 | 3 |
| Top Failure Mode | Other | Destructive Action |
Frequently Asked Questions
Is Claude or Claude Code more reliable?
Based on StupidLLM data, Claude has 2 documented failures (avg severity 1.8/10) while Claude Code has 8 (avg severity 6.4/10). Claude shows better reliability based on average severity scores.
What are the main differences between Claude and Claude Code failures?
Claude's most common failure mode is other, while Claude Code most commonly fails via destructive action. Claude has 0 critical incidents vs Claude Code's 3.