Today, there are only two options for using Google AI Studio and its Gemini 3.1 Pro model: pay-per-request or a monthly subscription. However, how can we justify paying for the platform when we still need to waste resources just to correct mistakes made by the AI? Sometimes I spend the whole day only fixing bugs and prompts that didn’t work out.
To address this, I strongly recommend that Google implements a solution: when a prompt is made to correct a mistake or a gap between the user’s need and the AI’s understanding, it should be free. We are not advancing; we are struggling, and that shouldn’t be paid for. There would be two consequences:
-
Users would be happy to only pay for real progress, not for training the AI.
-
Google would be encouraged to keep improving the model until this gap closes.
Both would benefit. But currently, it is unfortunately not a fair exchange.