Claude disruption: Anthropic services down for about 40 minutes
An outage hit Claude.ai, Claude Code and Claude Cowork on Saturday evening. Anthropic reported login failures from 21:58 UTC; the API stayed up.
Illustration · AI-generated (AI IN LIFE)
Anthropic confirmed a major disruption of its Claude services on Saturday evening.
At a glance
- Disruption began 21:58 UTC on August 16 (authentication issues)
- 22:07 UTC: Anthropic reported degraded performance on Claude.ai and platform.claude.com
- Affected: Claude.ai, Claude Code, Claude Cowork; API and Console stayed up
- Duration per reports: about 40 minutes until all-clear
- Root cause initially unclear – incidents "under investigation"
Anthropic confirmed a major disruption of its Claude services on Saturday evening. From 21:58 UTC, the status page reported authentication issues affecting Claude.ai, Claude Code and Claude Cowork; at 22:07 UTC it added "degraded performance" on Claude.ai and platform.claude.com. After roughly 40 minutes, operations were reportedly largely back to normal.
Anthropic initially described the incident tersely: it was investigating "an issue preventing some users from authenticating to Claude.ai, Claude Code, and Claude Cowork." Affected users reported failed logins, interfaces failing to load, and interrupted requests.
Important for businesses: the Claude API and Console remained operational according to Anthropic – production integrations via the API were not hit to the same degree as the consumer-facing products.
The company did not initially name a root cause; both incidents remained "under investigation" at the time of reporting, per BleepingComputer. The outage joins a series of brief disruptions at major AI providers this year.
The incident underlines how central AI assistants have become to daily workflows: even a sub-one-hour disruption triggered outage reports worldwide and renewed discussion about failover strategies.
FAQ
Which services were affected?
Claude.ai, Claude Code and Claude Cowork saw login and loading failures; the Claude API and Console remained operational per Anthropic.
How long did the outage last?
The status page reported issues from 21:58 UTC; after about 40 minutes, service was reportedly largely restored.
What caused it?
Anthropic did not initially disclose a cause. Both incidents were still under investigation at the time of reporting.
