Before it was: 100 o3 per week 100 o4-mini-high per day 300 o4-mini per day 50 4.5 per week
[0] https://help.openai.com/en/articles/11909943-gpt-5-in-chatgp...
I'm guessing they'll just announce massive tier generosity later considering how GPT-5 input tokens are half the price of 4.1 on the API. It's probably a way to keep the servers from being overloaded and to encourage people to buy Plus while the hype is hot.
The expensive thing: 100 per week -> 200 per week
This is...the opposite of a nerf? The numbers goed up? (We can quibble about the daily vs hourly difference, but certainly for me the weekly cap was the only thing that mattered.)
The better analogue is "Imagine in the 70's being able to teletype into an insanely expensive compute infrastructure and have reasonable timesharing capabilities of a limited resource across multiple users."
Unix. I'm describing the motivation for Unix there.
We already look back on earlier times with constraints that were appropriate.
Presumably compute will get cheaper, we'll build more datacenters, maybe we'll even power them in a way that doesn't destroy our planet, and GPT questions will become too cheap to meter. Just give it some time.