Anthropic flags connection resets on claude.ai’s Cowork; elevated UI errors under investigation
Published Mar 25, 2026, 3:22 PM UTC
Key entities
TLDR
Treat this as a claude.ai web UI transport instability centered on Cowork: keep API usage unchanged, add retry/backoff for UI-driven calls, and monitor the two active incidents for scope changes and resolution updates.
Why this matters
Scope likely UI-side: The pairing of a broad claude.ai errors notice with a more specific Cowork connection reset description implies the web UI/experience is principally affected, not underlying model inference or auth flows (medium confidence).
What changed
- Anthropic reported “Elevated Errors on claude.ai” with status Investigating and impact listed as None.
- A separate, concurrent incident cites “Elevated connection reset errors in Cowork,” also Investigating with impact labeled Minor.
- Both updates are recent and limited to claude.ai/Cowork; no mention of model degradation or API-wide outage in either note.
Topic context
Use this page when you want fast context on confirmed model launches from OpenAI, Anthropic, Google DeepMind, xAI, Meta, and similar labs without scanning every release note, model card, or developer post yourself. Key angles: openai, anthropic, google deepmind, gemini.
Summary
Anthropic posted two active incidents—general elevated errors on claude.ai and elevated connection reset errors in Cowork—both under investigation with minor/none stated impact, which together suggest a transport/UI-layer issue rather than a model outage or API-wide regression.