Breaking

OpenAI, Anthropic and SpaceXAI services disrupted in overlapping Thursday outages

September 3, 2026 at 22:29 EDT

Foundation Models
Infra & Chips
Software Dev & Coding

OpenAI, Anthropic PBC and SpaceXAI suffered overlapping service interruptions on Thursday, according to company status pages and user reports, disrupting customers of three of the leading U.S. AI model makers. Tens of thousands of users reported OpenAI problems on Downdetector, while Anthropic and SpaceXAI each logged elevated errors or outages on their own dashboards.

September 3, 2026 · AI infrastructure

Three labs, one overlapping outage window

ChatGPT, Claude, and Grok were disrupted the same Thursday morning. SpaceXAI later tied Grok’s failure to its Memphis compute center and apologized to unnamed compute partners.

Grok user reports (Eastern morning)

Vertical stacks are proportional to Downdetector counts: under 10 before 9 a.m. versus 1,365 at 9:45 a.m.

<10

Before 9 a.m.

1,365

Peak 9:45 a.m.

Takeaway: Grok complaint volume jumped more than a hundredfold in under an hour.

3.5 h

Grok down from 6:30 a.m. PT

99.63%

ChatGPT 90-day uptime

99.4%

Claude 90-day uptime

Incident timeline (Eastern)

9:23 a.m. Anthropic partial outage

→

10:43 a.m. OpenAI ChatGPT + Codex errors

→

12:16–12:55 p.m. Claude then OpenAI marked resolved

WHAT RECOVERED

By early afternoon, ChatGPT, Claude, and Grok were reported back. SpaceXAI said systems were functioning nominally after Memphis restored.

WHAT REMAINS UNCLEAR

Anthropic leases SpaceXAI compute, but OpenAI did not publish a shared root cause. Major cloud status pages did not declare matching incidents at press time.

Continue reading

The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.