BrokenAIClaude
Anthropic
- Founded
- 2021
- HQ
- San Francisco, CA
AI safety company founded by ex-OpenAI researchers Dario and Daniela Amodei. Builds the Claude family of large language models with a focus on steerability, interpretability, and reducing harmful outputs.
Products
Claude (Opus · Sonnet · Haiku) · Claude.ai · Claude Code · Anthropic API
Reliability — last 14d
Composite score
68.1/100
uptime · freq · sev · mttr
Uptime
97.60%
severity- and component-weighted
Incidents
20
top type: elevated errors
Median MTTR
46m
bad day: 3h 10m
What's breaking (last 14d)
Components ranked by incident count. Click a row for the incident list. Down-time is attributed proportionally for multi-component incidents.
| Component | Incidents | Down (attr.) | Median MTTR | Last |
|---|---|---|---|---|
claude.ai | 14 | 4h 1m | 54m | 1d ago |
Claude Code | 13 | 4h 1m | 55m | 1d ago |
Claude Cowork | 13 | 5h 23m | 55m | 1d ago |
platform.claude.com (formerly console.anthropic.com) | 13 | 3h 54m | 53m | 1d ago |
Claude API (api.anthropic.com) | 12 | 3h 38m | 54m | 1d ago |
Claude for Government | 4 | 1h 22m | 56m | 1d ago |
Unspecified | 2 | 18m | 9m | 8d ago |
Type of failure
Categorised from the public incident headlines (outage / latency / auth / etc.). Click a category to drill in.
When it breaks
Incident starts by weekday × hour, UTC. Hover a cell for incident detail.
S
M
3
T
2
W
3
T
5
F
4
S
1
0006121823
Hour of day · UTC
Anthropic — incidents by day (last 14d)
MinorMajorCritical
Total 18 incidents
1d incident-free streak
Worst incidents
Ranked by severity-weighted duration (impact × wall-clock).
- #1Elevated errors on Claude.ai, API, Claude Codecritical3d agoduration: 2h 50mtype: Elevated errorsaffects 4 components
- #2Elevated errors on Claude.aicritical12d agoduration: 1h 27mtype: Elevated errorsaffects 4 components
- #3Sonnet 4.6 elevated rate of errorsmajor11d agoduration: 3h 10mtype: Elevated errorsaffects 5 components
- #4
Score breakdown
Uptime50%
76.0
Frequency20%
75.1
Severity20%
25.6
MTTR10%
99.6
SLA compliance
How often does this provider clear their stated availability target?
Anthropic Commercial Terms · 99.90% · source ↗
Met SLA in 0 of 13 weeks (0%)✗ last 14d97.60%-2.30%
Incident history
showing 6 of 200 · 4 of 105 days
Fri, Apr 172026-04-171 incident
Thu, Apr 162026-04-163 incidents
Wed, Apr 152026-04-151 incident
Tue, Apr 142026-04-141 incident