Anthropic’s Claude service ran into problems Monday, March 2, 2026, with users reporting trouble logging into the chatbot and developer tools. “Claude is currently experiencing a temporary service disruption. We're working on it, please check back soon,” an error message read.
On its public status page, Anthropic said that it was investigating the issue, adding that “multiple services affected, including claude.ai, platform.claude.com, the Claude API (api.anthropic.com), Claude Code, and Claude for Government,” have been affected.
Users said on social media that they were unable to log in to claude.ai and were met with authentication errors or failed responses inside the web app. Some said their sessions expired and they were unable to sign back in. Developers also reported API requests failing and errors appearing inside Claude Code and the Anthropic console.
This comes hours after Amazon Web Services Inc. (AWS), Amazon's cloud unit, said that its UAE-based data centre had been struck by "objects," sparking a fire and was temporarily shut down. The company also said that it was investigating connectivity and power issues at a separate zone of data centres in Bahrain that had been impacted by a "localised power issue."
The company did not say whether or not this was caused by the air strikes from the ongoing war between the United States and Israel with Iran. Both Dubai and Bahrain have been hit by air strikes during the war.
On the main status dashboard, claude.ai, platform.claude.com, the Claude API, and Claude Code were marked “Partial Outage.” Claude for Government was listed as “Operational.”
The first status entry appeared at 11:49 UTC and was marked “Investigating,” stating that the company was looking into the issue. At 12:06 UTC, Anthropic posted that it was continuing to investigate.
At 12:21 UTC, the company wrote that the Claude API was “working as intended” and that the issues it was seeing were related to claude.ai and the login and logout paths.
At 13:22 UTC, the incident moved to “Identified,” with Anthropic saying a fix was being implemented.
At 13:37 UTC, the company added another update stating that some API methods were not working and that it was investigating.
At 14:05 UTC, the latest update read: “We are continuing to work on a fix for this issue.”
As of the 14:05 UTC entry, the incident remains open. The status page does not show a resolved notice, a cause, or an estimated time for restoration.
The dashboard continues to display component-level availability for each Claude service and links to uptime history.
Damilare Odedina
