Claude suffers repeat outage
Users reported another outage of Anthropic’s Claude AI this month — the third Claude disruption in recent weeks, with login failures and slow responses reported. Recurrent platform instability is a reminder that critical workflows depending on third‑party AI need documented fallbacks and cross‑trained backups.
Anthropic flagged a major incident on March 2, 2026 at 11:30 UTC (bleepingcomputer.com) that kicked off the first widely reported service disruption this month. Independent trackers showed a sharp surge in reports during that March 2 window, with one aggregation citing nearly 2,000 user complaints at peak on Downdetector-style feeds. (dataconomy.com) Anthropic’s own status feed labeled the disruptions as “Elevated errors on claude.ai” and records a separate, short incident on Mar 12 that ran roughly 8:30–8:44 PT. (anthropic.statuspage.io) Multiple outlets and incident trackers noted the core Claude API continued handling traffic while user-facing frontends and developer tooling were the most visible error sources, according to reporting and post‑incident summaries. (ibtimes.com.au) On Mar 17, real‑time outage trackers logged a fresh wave of user reports—Business Today recorded about 120 Downdetector flags during that event—as Anthropic posted that it had identified the cause and was rolling out a fix. (businesstoday.in) Community monitors such as IsDown showed recurring elevated‑error reports (83 on Mar 16 in one feed), reinforcing that multiple, short incidents populated the month’s timeline and that status.claude.com remained the canonical incident timeline for teams tracking service health. (isdown.app)
