OpenAI has announced that starting in December, usage quotas across its three paid tiers—Plus, Pro, and Business—will be shared between ChatGPT and Codex and measured uniformly by token, with only Instant chat exempt from quota limits. Once a user's quota is exhausted, they can choose to wait for a reset, upgrade their plan, or purchase credits to keep using the service. At the same time, the new rules tighten the cap on the number of paid accounts that can be logged in on a single device; exceeding that limit will result in restricted login. Previously, Codex lead Tibo completed 28 consecutive days of updates, during which quotas were reset multiple times and numerous product improvements were rolled out. This adjustment marks a shift by OpenAI—from frequently resetting quotas to a token-based billing model that passes computing power costs on to users more explicitly. Heavy users will be the first to feel the squeeze on their quotas, and the community reaction has been rather negative.