Three of the world’s leading artificial intelligence platforms, ChatGPT, Claude and Grok, suffered significant outages within hours of one another on Thursday September 3, 2026, disrupting users worldwide whilst raising fresh concerns about the reliability of the infrastructure supporting the rapidly expanding AI industry.
OpenAI’s ChatGPT, Anthropic’s Claude and xAI’s Grok all experienced service disruptions on Thursday September 3, 2026, with users reporting errors and difficulties accessing the platforms. The incidents occurred close enough together to spark speculation over whether the services were affected by a common infrastructure problem, although none of the companies has confirmed a shared cause.
OpenAI attributed the disruption affecting ChatGPT and its Codex coding service to a routing error. The company said it had implemented a fix and later confirmed that the incident had been resolved.
Anthropic separately reported a partial outage affecting several Claude models and services, including Claude Code and its API. The company attributed the disruption to an infrastructure issue and reassured the public that the affected services had been restored.
Meanwhile, xAI said the problems affecting Grok were caused by an outage at its Memphis computer centre. The company later reported that traffic had returned to normal.
Earlier reports suggested that problems involving Microsoft Azure could have even responsible for the disruptions because the cloud provider has infrastructure relationships with major AI companies.
These disruptions prompted a sharp increase in user reports on outage-tracking platform Downdetector. Reports concerning ChatGPT in the United States exceeded 35,000 at the height of the disruption, while Claude and Grok each recorded more than 1,000 reports.
These near-simultaneous outages were unusual because ChatGPT, Claude and Grok are competing AI services operated by separate companies. Their close timing underscored a broader vulnerability: as AI systems have become critical digital infrastructure, with platforms increasingly embedded in everyday activities, disruptions affecting any individual platform can be easily felt by users and businesses around the world.
This highlights the growing dependence of the AI industry on a complex network of data centres, computing infrastructure, cloud services and internet-routing systems. A single disruption has the potential to interrupt coding projects, research, customer-service operations and other workflows that now rely on AI tools. It also demonstrated the limits of simply switching between competing AI assistants during a major disruption. Although Google’s Gemini did not report an official outage during this downtime, some users were found to have reported difficulties using the service.
All three AI chatbots are yet to disclose the cause for this major downtime.
