What happened to intelligence "too cheap to meter"?
retained text
I've noticed anecdotally here on reddit ( and myself ) that OpenAI has been reducing the amount of tokens that people get with their plans ( models like 6.1 eating through budgets way faster ) and is implementing putting a cap on reasoning in chatGPT in December. What happened to Sam Altman's goal of intelligence "too cheap to meter"? Is there a supply chain reason the companies are gatekeeping intelligence? Is there any internal rumors of more token efficient models maybe? Or is it just greed?
document fetched 2026-11-02 00:27 · canonical https://reddit.com/r/OpenAI/comments/1xlensq/what_happened_to_intelligence_too_cheap_to_meter