.png){kind=logo}
OpenAI Shifts ChatGPT Free Users to Cheaper Default AI Model
OpenAI has quietly rewired how ChatGPT works for users on its free tier and low-cost Go subscription. The tech giant has shifted them to a more affordable AI model by default. The move, as cited by Wired, suggests the company is reworking both user experience and costs as it scales its services.
Under this new arrangement, prompts coming from free and Go users are now processed by the GPT-5.2 instant model. OpenAI has also ceased automatically switching over certain queries to more advanced reasoning-focused models in the background.
What Has Actually Changed for Users?
ChatGPT automatically routed complex or sensitive prompts to a more capable ‘Thinking’ model when it believed deeper reasoning was required. The process happens behind the scenes, where the user’s choice is not required. OpenAI has now ceased this service to give users more control.
The company announced this update on December 11. Although GPT-5.2 Instant is now the default model, free and Go plan users can still manually switch back to thinking using the tools menu while composing a prompt. However, the switch must be made intentionally and repeatedly.
GPT-5.2 Instant vs Thinking model
OpenAI characterizes GPT-5.2 Instant as a general-purpose model suited for everyday tasks, learning, and fast responses. The Thinking model, on the other hand, has been optimized for difficult problems that require deeper reasoning, step-by-step analysis, and refinement.
While it does not entirely remove access to the advanced model for free or Go users, this new system no longer applies it automatically, even for complex problems.
Also Read: Meta Goes AI-First: Employees Get ChatGPT-5, Gemini Access
Responding to User Backlash, and Cutting Costs
This update has begun facing public scrutiny over its automated model selection. After the GPT-5 launch, several users claimed auto routing has reduced the personality and quality of ChatGPT’s responses.
Even the CEO of OpenAI, Sam Altman, expressed dissatisfaction with the way this system works.
However, the shift likely delivers clear cost savings. Defaulting millions of free users to a cheaper model could significantly reduce compute expenses, especially if consumers do not switch models manually.
Questions of Safety and Changes Future
The update also comes with potential safety implications. OpenAI previously said sensitive conversations, including mental health-related queries, were automatically handled by reasoning-focused models.
With that safeguard removed, GPT-5.2 Instant has been improved to handle such cases, although its practical effectiveness is not yet proven. The company has also not clarified if free and Go users will experience further usage restrictions.
