Anthropic Stays Silent as AI Rivals Explain Outages — SkimNews
Get the Tech newsletter
Daily tech — startups, AI labs, chips, the launches that shape the next decade. Free.
- Anthropic, OpenAI, and xAI all experienced rare outages Thursday morning (Sept 3); OpenAI cited a routing error starting at 7:43 am PT and resolving by 8:17 am PT, while SpaceX told Grok users the issue was "an outage at our Memphis compute center."
- Anthropic alone declined to explain its partial outage, which began at 6:23 am PT and affected Claude Mythos 5.1, Claude Fable 5.1, and Claude Opus 5 before being marked resolved by 9:16 am PT, with Claude Sonnet 5 briefly affected shortly after 9 am PT.
- xAI's Grok outages began at 6:30 am PT and were resolved at 10:05 am PT; SpaceX, xAI's parent company, also publicly apologized to "impacted compute partners."
- Anthropic and xAI both announced a "compute partnership" with SpaceX in May — context the article flags as relevant given xAI blamed its outage on a SpaceX-operated facility.
- OpenAI spokesperson Kathleen Chaykowski told WIRED the routing error made ChatGPT and Codex unavailable "for some users across platforms."
- Cloudflare, Amazon Web Services, and Microsoft Azure all reported no outages Thursday, eliminating the typical shared-vendor explanation for simultaneous sector-wide incidents.
- Google did not confirm scattered reports of a possible Gemini outage Thursday morning and recorded no incident on its service status dashboard.
Why it matters: Anthropic's silence stands out because it and xAI both partnered with SpaceX for compute in May, raising the question of whether Anthropic's outage shares the same infrastructure root cause as Grok's Memphis data center failure. With Cloudflare, AWS, and Azure all confirming no issues, the usual shared-vendor explanation doesn't apply, leaving enterprise customers without clarity on whether the incidents are truly independent.
Ask SkimNews
