Unfortunately, they reserve the right to change the terms after the fact.
What bothers me about these cuts is the assumption that because the models are supposedly getting better and cheaper, less usage is somehow equivalent value. That may be true from their perspective, but they don't get to decide how much utility those extra capabilities have for every user.
It also leads to the AI version of "you’re holding it wrong" (i.e. you’re prompting it wrong/use a cheaper model/skill issue).
And I can't help wondering how much of this is also about shifting capacity toward the latest persistent-agent offering they're pushing. If that's the case, existing subscribers are effectively being asked to give up quota for a product direction they may not even care about.
Been a customer since GPT 5.3 Codex. This is the first time that I'm thinking of dropping the sub.