I hear a lot of people say that it keeps breaking on them, and they spend hours trying to fix it.
I have not faced these issues in a long time (I stopped updating it and put up a lot of guardrails and recovery scripts) but I am curious what you guys are trying to do with it.
When Claude Opus got lobotomized their fanboys and what I consider cultists piled on any criticism and just call it skill issue.
There is a certain emotional attachment there in their sub towards a tool that I really think is pathetic, and it’s a breath of fresh air to see the opposite here. I don’t see half the level of gaslighting here as I witness by the cultists over there at the other sub.
Loyalty and emotional attachment don’t make any sense. If a tool is getting worse and worse the right choice is to criticize it and then to move on to better tools, not act like literal cultists and shills for said company. Brand loyalty makes 0 sense. Once OpenAI starts behaving like Anthropic, everyone should jump ship.
Just wanted to put this out here.
Cancel your codex subscription, go subscribe to Claude code. You’ll get shit on by Anthropic relentlessly, just the way you like it, and you’ll get zero resets, just the way you like it. That way you won’t have to deal with openAI being “predatory” and “abusive” by resetting the usage limits.