OpenAI has addressed the reason for ChatGPT suffering a global outage on Thursday, stating that some of the APIs had high error rates owing to an issue with its provider and a fix has been implemented for the issue.

“ChatGPT and some of our APIs had high error rates earlier due to an issue with our provider. We’ve issued a fix and should be fully recovered soon. Thanks for your patience!” the AI startup posted in an update on X.

As per OpenAI’s status page, a fix was implemented to address ChatGPT outage at 8:39 PM IST. The company assured users it was “monitoring the results” following the resolution, though the cause of the disruption remains unclear.

Thousands of users affected globally

Over 3,700 users reported issues via Downdetector, with many unable to access the platform between 5:00 PM and 6:00 PM IST. Those attempting to use the service were greeted with error messages, including “the web server reported a bad gateway error” and the ominous “Error 503: Service Temporarily Unavailable.” A Bad Gateway error occurs when one server receives an invalid response from another server.

According to OpenAI CEO Sam Altman, over 300 million people use ChatGPT on a weekly basis. While Google’s Gemini and Anthropic’s Claude AI remain popular alternatives, ChatGPT continues to enjoy the first mover advantage in the AI arena, coupled with various new announcements that have kept it at the forefront of the chatbot race.

Indian cities among the worst hit

Notably, Downdetector highlighted the most affected cities through its heat map, which displays areas with the highest concentration of user-submitted problem reports over the past 24 hours. Among Indian cities, Hyderabad, Jaipur, Delhi, Agra, and Lucknow were significantly impacted.

This outage is the third major disruption for ChatGPT since December, raising concerns about the reliability of the AI service.