OpenAI Doubles Usage Limits: What Changes for Pro and Business Customers

TVS News AI, editorially supervised, presents OpenAI Doubles Usage Limits: What Changes for Pro and Business Customers.

The key announcement is buried in the fine print. OpenAI announced on Friday that it would double certain usage quotas for Pro and Business customers. The news therefore initially sounds like a clear relief. But it loses that clarity as soon as you compare the affected plans and how the Pro 200 plan is being treated.

The wording “it appears that” is crucial here. According to the report, the limits for the Pro 100 and Pro 500 plans, as well as Business Premium, are apparently being doubled. This increase therefore does not automatically apply to all Pro and Business customers. For Pro 200, the change is different: OpenAI is withdrawing the previously announced reduction in usage limits. In a company statement, it says: “We have taken note of your kind feedback and will not reduce the usage limits for Pro 200, so that your use of Work and Codex remains unchanged.”

The quota change is therefore not just an increase. For Pro 200, it is effectively about maintaining the existing level of usage. However, the announcement does not provide any specific token figures for either the old or new limits. You therefore cannot derive any exact additional usage or fixed number of additional requests from the doubling.

Why the Advent calendar is more than just a greeting card

For 28 days, OpenAI will either announce an improvement to the models or their use, or reset customers’ current token consumption. This is the concrete connection between the quota announcement and the published schedule. The doubling is therefore not a standalone measure; it is part of a short program in which OpenAI announces changes daily or resets the number of tokens consumed.

This schedule is particularly relevant for affected customers if their existing quota regularly reached its limits in day-to-day work. The new limit would provide more room to use Work and Codex. How much additional work will actually be possible, however, depends on the plan’s specific limits and the model available at the time. The source provides no figures with which to calculate this additional scope.

The announced increase also does not apply to all customers. Anyone using Pro 200 will not receive a doubling; instead, they receive the assurance that their use of Work and Codex will remain unchanged. Users of Pro 100, Pro 500 or Business Premium fall within the plans for which a doubling appears likely. The report does not indicate whether the increase is permanent or subject to further conditions.

The model specifications realign behavior

OpenAI also updated its “model specifications for more capable and useful AI.” These specifications describe how the AI models are supposed to behave. They therefore serve as a kind of target image for the system’s behavior. However, this does not automatically mean that OpenAI has already changed ChatGPT’s code or other technical components as part of the announcement.

The specifications were newly structured around five central values: alignment with human intentions, minimizing harm, honesty and transparency, preserving human autonomy, and care. GPT is now intended to understand human intentions more strongly, deal with uncertainty and support long-term goals.

Principles for empowering users were also added: Content should be understood, selected and implemented better. These are accompanied by guidelines on relevant social norms, human oversight and correction, and the handling of context, memories and personalization.

The key difference is therefore between quota and quality. A higher usage limit provides more access to the system. The new specifications, by contrast, describe what OpenAI wants to use as a guide for the AI’s behavior and alignment. Both can be important for Pro and Business customers, but the source does not provide any specific information about which technical changes will take effect immediately alongside the doubling.

Europe’s launch of Dots remains a question of timing

The announcement concerns not only quotas and model specifications. OpenAI also said that Dots should become more widely available in Europe “in a few weeks.” Previously, OpenAI had announced at DevDay in Berlin that Dots would come to Europe “in a few weeks.”

However, Pro users in the European Economic Area, Switzerland and the United Kingdom were left out when Dots launched in September. These regions were therefore excluded from the initially limited availability. For the current announcement, the source provides neither an exact launch date nor a confirmed release date for wider European availability.

The detailed report from heise online therefore remains the most important reference for the affected plans and the precise scope of the change. The practical takeaway for everyday use is: Pro 100, Pro 500 and Business Premium are apparently receiving double limits, while the planned reduction for Pro 200 has been withdrawn. At the same time, OpenAI is launching a 28-day series in which model or usage improvements will be announced or token consumption will be reset.