On Thursday afternoon, OpenAI’s popular AI chatbot, ChatGPT, encountered a significant disruption, leaving users unable to access its services. Starting at around 1:30 PM ET, users began reporting issues such as an inability to process queries and “internal server error” messages.
The outage quickly gained traction on monitoring platforms like Down Detector, indicating a widespread impact.
At 2:00 PM ET, OpenAI confirmed the issue on its status page, reporting high error rates across ChatGPT, its API, and its newly released text-to-video generator, Sora. By 6:15 PM ET, OpenAI shared that Sora had been fully restored and APIs were beginning to recover.
However, the company noted ongoing efforts to address broader functionality problems with ChatGPT and its APIs.
“This issue is caused by an upstream provider and we are currently monitoring,” said the ChatGPT outage status. “We are currently experiencing an issue with high error rates on ChatGPT, the API, and Sora. We are currently investigating and will post an update as soon as we are able.”
While OpenAI did not directly identify the cause, it hinted at complications related to an “upstream provider.” Microsoft, OpenAI’s exclusive cloud partner, reported a coinciding “power issue” at one of its North American data centres, which may have contributed to the disruption, reports The Verge.
The issue also impacted Microsoft’s Xbox cloud gaming services.
By 5:00 PM ET, Microsoft announced it had “fully restored” power to the affected data centre. This timeline aligns closely with the gradual recovery of OpenAI’s systems, though the chatbot’s complete functionality was not immediately restored.
Thursday’s incident is part of a growing list of service interruptions for ChatGPT. Earlier this month, ChatGPT and Sora experienced a prolonged outage just days after Sora’s launch to ChatGPT subscribers. A major outage affecting multiple AI tools brought ChatGPT offline in June.
In the News: General Dynamics employees targeted in phishing attack
