Hi! I have a product idea for OpenAI that I’d like to pass along to the product team.
Idea: Traction-Based AI Compute Grants for Builders
Many people have good product ideas but cannot afford the AI compute needed to build them.
What if OpenAI offered two builder modes?
Free Builder
- A limited amount of advanced coding-agent compute is provided for free.
- The project must display an attribution such as “Built with ChatGPT”.
- If the project demonstrates genuine traction, its free compute allowance increases.
Traction could be measured using multiple signals rather than GitHub stars alone: active users, retention, forks, contributors, product usage, growth, community activity, etc.
Paid Builder
- The user pays normally.
- No OpenAI attribution is required.
- They receive higher limits and private/professional usage.
This creates a mutually beneficial loop:
Free compute → build a product → gain real users → demonstrate traction → receive more compute → grow → eventually become a paying customer.
For OpenAI, this could function as both a developer-acquisition program and an AI-native startup incubator. Successful products would organically showcase what OpenAI can build, while unsuccessful projects would not receive unlimited additional compute.
The key idea is not simply “make AI free.”
It is:
Use controlled AI compute as a performance-based grant for promising builders.
I think this could be worth testing as a small pilot program.
Please forward this suggestion to the appropriate product/business team if possible.