OpenAI is merging its Chat and Work features into a single interface in ChatGPT, introducing a shared token budget starting in December for Plus, Pro, and Business plans. This update folds Work's agentic features directly into the main platform while combining usage limits that previously governed ChatGPT and Codex. The change effectively signals a shift away from simple all-you-can-prompt pricing as users consume a single token pool for various tasks.
Because tokens measure the information a model processes and generates, more complex models and tasks will deplete the budget at a faster rate. Users can monitor their remaining balance and reset times directly within Settings. Once a budget limit is reached, individuals have the options to wait for a reset, upgrade their subscription, or purchase extra credits to continue. While chats utilizing Instant models remain exempt from the budget, heavy usage on cheaper tiers will likely result in lower-quality outcomes.
This restructuring follows OpenAI's recent doubling of usage limits for higher-priced plans in preparation for the merger. The consolidation also addresses growing criticism regarding ChatGPT's increasing complexity, as users previously had to navigate between multiple operating modes and numerous model settings. Competitors like Anthropic have similarly combined their chat and coworker features, pointing to a broader industry trend toward unified AI workspaces.
Why it matters: This transition away from flat-rate pricing models marks a significant shift in how everyday users consume and pay for advanced artificial intelligence capabilities.