In the past few days, a batch of similar posts have appeared in the Reddit and OpenAI developer communities: the ChatGPT account was suddenly deactivated, and when opened, only a line of error code account_deactivated was seen.
There are quite a few of them who are Pro users for $200 per month. On November 7, someone posted that his Pro account was blocked directly that day, without receiving any warning or ban notification email.
The most devastating thing is not being blocked, but being unable to get out the things inside. Long-term accumulated conversation records, project information, and customized GPTs will all be locked as soon as the account is stopped. Some people say that almost all of their workflow is based on it.
Someone on Reddit has been compiling recent cases, and the reasons appear to be diverse, involving different types of usage policy disputes, but there is no unified statement. So far OpenAI has not come out with an explanation.
This kind of no-warning account ban has appeared in various AI products. Common trigger points include frequent logins from different places, multiple people sharing accounts, abnormal payment methods, or a certain conversation being misjudged by risk control. But the biggest dissatisfaction of users this time is "not telling the reason": they don't know which line they have stepped on, and there is no way to appeal.
If you rely heavily on ChatGPT for work, it’s best to export your data regularly now. There is an export entry in the settings, which will be packaged and sent to you via email.
Paying does not mean exemption from inspection. This time taught many people a lesson.