Simultaneous outages hit ChatGPT, Claude, Grok, and Gemini across global networks

Simultaneous disruption across prominent platforms

On Thursday, 3 September 2026, leading artificial intelligence platforms experienced simultaneous outages and degraded performance worldwide. The incident began around 10:30 AM Eastern Time (16:30 Central European Time) and affected OpenAI's ChatGPT, Anthropic's Claude, xAI's Grok, and Google's Gemini. In Spain, monitoring tracked severe accessibility problems beginning at 15:00 peninsular time and peaking by 17:00, preventing users from logging into accounts or receiving query outputs. Users across Europe and the Americas reported persistent error prompts, connection timeouts, and failures when submitting text queries or attempting image generation on web and mobile interfaces. Monitoring platform Downdetector registered steep concurrent spikes in error reports for each provider starting in the afternoon, while Perplexity operated without disruption.

Scope of impact across model tiers

Technical dashboards revealed that the disruption touched core computing layers across several competing AI developers. OpenAI designated the event as degraded performance on its official status page, confirming elevated error rates across its primary chatbot and developer tools. Status monitoring showed that 15 separate components of ChatGPT and 4 components of Codex were affected during the incident. Anthropic noted that the disruption initially hit multiple model versions, specifically Mythos 5.1, Fable 5.1, Opus 5, Opus 4.8, and Opus 4.6. As engineering teams applied fixes, Anthropic narrowed the active service disruption to its Opus 4.8 and Opus 5 models.

Remediation efforts and provider statements

Engineering teams deployed emergency patches to restore functionality for millions of daily users, enterprise clients, and external developers. Anthropic confirmed that the disruption stemmed from an infrastructure failure across its consumer, enterprise, and developer environments.

Claude is mostly back up after an infrastructure issue caused a partial outage across Claude.ai, Claude Code, Claude Cowork, and the Claude API. We will share updates on our status page as we continue to restore service.

Anthropic also acknowledged the disruption to daily workplace operations, noting that Claude forms a core part of how users complete professional tasks. OpenAI also enacted mitigation protocols shortly after error counts surged, with representatives pointing to its active status monitor.

We have applied the mitigation and are monitoring the recovery.

Underlying infrastructure and recovery tracking

The disruption temporarily halted productivity workflows across global businesses and developer environments that rely on automated API calls for software tasks. xAI confirmed that Grok experienced service disruptions across its web interface and mobile application, noting that staff were working to restore normal access. Because multiple independent AI platforms faltered in the same hour, initial technical assessments pointed toward shared dependencies in large cloud hosting platforms, backbone internet routing, or providers such as Microsoft Azure. OpenAI released a brief video update on X as ChatGPT returned to operational levels, while monitoring continued across all affected systems to ensure stability.