ChatGPT suffered a sudden large-scale outage and millions of users were unable to log in. The focus in the field of artificial intelligence has recently focused on server stability again.

ChatGPT suffered a sudden large-scale outage, and millions of users were unable to log in. The focus in the field of artificial intelligence has once again focused on server stability. According to media reports, the world-renowned AI conversation application ChatGPT suffered a sudden large-scale global service interruption around August 20. This failure affected a wide range of areas. Millions of users in the United States, Europe and other regions found that not only were they unable to log in to their accounts or register new accounts normally, they were also faced with a severe situation where the chat interface was completely stuck and historical conversation records could not be read. Judging from the actual feedback and fault performance of users, this outage has had a comprehensive impact on daily interactions. When many users try to access, they will find that the sidebar of the application is always stuck in the loading animation state, and they will continue to receive error messages caused by "too many concurrent requests" when sending messages. In addition, the core verification and account channels of the ChatGPT.com official website are also paralyzed. In addition to the consumer-oriented chat interface, OpenAI's developer code platform Codex and as many as 12 core API interfaces were not immune, with large-scale operation abnormalities and high error rates occurring one after another. Faced with the sudden global crisis of trust, OpenAI officially responded quickly and immediately announced the details of the failure on its status page. The official technical team confirmed that all affected services are focused on user login and verification abnormalities, and made it clear that they are currently fully promoting the implementation of emergency repair plans. As of shortly after the failure occurred, officials had marked the incident status as "problem has been located and is still being processed." With the subsequent repair work fully completed, various core services and API interfaces are gradually returning to normal. This incident has once again sounded the alarm for the stability of AI infrastructure under high-load operation. via AI News (author: AI Base)