Elevated error rates on requests to some models

S status.claude.com ↗

▲ 58 points • 56 comments • by rob • 1w ago • HN discussion ↗

Pangram verdict · v3.3

We believe that this document is fully human-written

4 %

AI likelihood · overall

Human

100% human-written 0% AI-generated

SEGMENTS · HUMAN 1 of 1

SEGMENTS · AI 0 of 1

WORD COUNT 84

PEAK AI % 4% · §1

Analyzed

May 15

backend: pangram/v3.3

Segments scanned

1 windows

avg 84 words each

Distribution

100 / 0%

human / AI fraction

Verdict

Human

Pangram v3.3

Article text · 84 words · 1 segments analyzed

Human AI-generated

§1 Human · 4%

Incident

Report for ClaudeSubscribe to Updates Identified We are currently working to resolve issues affecting requests to Claude Opus and Sonnet 4.6, though success rates for Opus 4.7 have returned to normal. Posted May 15, 2026 - 01:06 UTC Update We are continuing to investigate this issue. Posted May 15, 2026 - 00:19 UTC Investigating We are currently investigating this issue. Posted May 15, 2026 - 00:18 UTC This incident affects: Claude API (api.anthropic.com) and Claude Code. ← Current Status Powered by Atlassian Statuspage