OpenAI Responds to Codex Usage Limit Discrepancies, Cites sub2api Abuse as Cause

On August 21, OpenAI responded to discrepancies in Codex usage limits, stating it will not make unauthorized changes without first communicating with the community.

An official investigation found that affected users largely converted subscriptions into API traffic via sub2api and redistributed it, a behavior flagged by fraud prevention systems. OpenAI noted that normal subscription use via Sign in With ChatGPT on official or OSS clients (such as Pi, OpenCode) is entirely fine.

[TechFlow]