Access to Anthropic’s Claude has been restored after the AI chatbot experienced disruptions in several regions worldwide for the second time in under 24 hours, data from Downdetector.com shows.

Anthropic’s latest large language model, Claude Opus 4.6, along with its AI tools including Claude CoWork and Claude Code, recorded “elevated errors,” according to the company’s official status page.

The issue has since been resolved, with an update posted at 7:59pm IST on Tuesday, March 3.

Claude had also suffered a major outage on Monday, March 2, when more than 900 users reported problems at the peak of the disruption around 5:30pm.

Anthropic said it was investigating “elevated errors” affecting the Claude chatbot, Claude Opus 4.6, its latest large language model, as well as Claude Console, the developer platform, and Claude Code, its AI-powered coding assistant.

The update was posted on the company’s status page at 5:19pm IST on March 2.

Claude’s recent outages appear to have started on the same day that Amazon Web Services reported disruptions after one of its data centers in the UAE was hit by “objects,” reportedly causing a fire.

The UAE is currently dealing with the aftermath of ⁠retaliatory missile and drone strikes by Iran following US and Israeli attacks on Iranian targets.

These Iranian strikes have affected airports, ports, residential areas, and other sites across the country and the broader Gulf region.

Two of Amazon Web Services’s cloud zones — clusters of data centers, in the UAE were reported to be without power on Monday, according to the company’s status page.

AWS serves as Anthropic’s primary cloud provider and training partner, though there is no confirmed link between the two incidents.