Around 9:23 a.m. Eastern Time on September 3, 2026, a large-scale failure occurred in leading AI services in the United States. According to Down Detector monitoring data, Anthropic’s Claude series of models first reported errors, followed by xAI’s Grok being unavailable across all platforms. Approximately one and a half hours later, OpenAI’s ChatGPT and Codex also experienced access issues. The incident lasted about 3 hours and 40 minutes, with over 37,000 error reports regarding OpenAI services worldwide during the peak period, 80% of which were related to ChatGPT. During the same period, Google Gemini and Microsoft Copilot also suffered impacts. On the evening of September 3, the Zhipu X platform issued a statement indicating that its services were operating normally.
# Three major AI models overseas experienced a near-four-hour outage! Zhipu X platform announced: We’re still online
On the evening of September 3, all leading AI services in the United States experienced failures, including ChatGPT, Claude, Grok, etc. According to statistics, this large-scale outage lasted approximately 3 hours and 40 minutes, with the failures first occurring around 9:23 a.m. on September 3, Eastern Time.
According to data from the overseas fault monitoring platform Down Detector, Anthropic’s Claude series of models saw a sudden increase in error rates, including core models such as Mythos 5.1, Opus, and Fable 5.1. Just two minutes later, xAI’s Grok service went down entirely; about one and a half hours later, OpenAI’s ChatGPT and programming tool Codex also experienced large-scale access errors. During the peak of the incident, there were over 37,000 fault reports related to OpenAI services worldwide, with 80% of them involving ChatGPT. During the same period, Google Gemini and Microsoft Copilot also…