Open ChatGPT or Claude right now. Look at the top of the screen. Which model is selected?
Almost certainly the biggest, most expensive one. The platform defaults you there. It doesn’t ask. It doesn’t explain. It just assumes you want maximum firepower for every single thing you type.
That’s fine when you’re doing serious work. Drafting a strategy document. Analysing something complex. Writing something that needs to be genuinely good. Use the best model you have access to.
But most of what people ask AI to do isn’t that. It’s a quick email reply. A bullet list. A summary of something short. A yes or no with a brief reason. For that kind of work, the cheaper, faster models are just as capable. They’ll give you the same answer in half the time.
The fix
Every major AI platform has a model selector. It’s usually a small dropdown near the top of the chat window. Switch it to the lighter option — GPT-4o mini, Claude Haiku, Gemini Flash, whatever the “smaller” version is called. Use that for everyday tasks.
Save the expensive model for when you actually need it.
The payoff
You’re probably burning through your monthly limit on tasks that didn’t need it. One click fixes that.
