10.06.2025 18:27
OpenAI acknowledged a significant service disruption affecting its popular AI chatbot, ChatGPT, along with other services, on Tuesday. Initially described as a partial outage, the problem, which began around 3 a.m. EDT, caused widespread user issues, characterized by elevated error rates and significant latency across multiple OpenAI platforms, including their application programming interfaces (APIs) and the AI image generator, Sora.
According to reports compiled by DownDetector.com, over 1,900 users experienced problems during the peak of the outage, with the vast majority (92%) specifically citing difficulties with ChatGPT. While the number of reported issues substantially decreased by midday, OpenAI maintained that degraded performance persisted even after implementing a mitigation strategy. The company attributed the issue to identified root causes and claimed to be actively working on comprehensive solutions.
By late morning, OpenAI announced that they had pinpointed the source of the problems. A temporary fix was deployed around 11:42 a.m., leading to improved performance for both the API and ChatGPT. However, even with the implementation of a more substantial mitigation at 1:20 p.m., OpenAI refrained from declaring a full resolution, instead characterizing the lingering challenges as "degraded performance." Despite multiple attempts, OpenAI did not provide further comment on the specifics of the situation or the long-term solutions they are pursuing.
This latest incident adds to a growing list of recent service disruptions for ChatGPT. Similar outages, marked by high error rates and impacting both ChatGPT and other OpenAI products such as Sora, occurred in March and January. These earlier events, while also significant, were resolved more swiftly. The sheer scale of the user base—estimated at over 500 million weekly active users as of March 31—underscores the potentially far-reaching consequences of such disruptions. The frequency of these outages raises concerns about the stability and resilience of OpenAI's infrastructure, particularly given the increasingly heavy reliance on its services across various sectors.
