OpenAI Tipped to Be Working on a Clinician Mode and ‘Speak First’ Voice Feature for ChatGPT

OpenAI has been shipping new features at a breakneck pace over the last few months. Recently, the San Francisco-based artificial intelligence (AI) giant launched the Sora app for iOS and announced third-party app connectivity within ChatGPT. Now, a tipster claims to have found evidence of a new “clinician mode” in the Chatbot, which could be a specialised healthcare or medical discussion feature. Interestingly, the AI firm is said to be developing a specific “model speaks first” prompt for voice chats.

OpenAI Could Soon Offer a Clinician Mode

Tibor Blaho, Lead Engineer at AIPRM, shared new strings of code from the web version of ChatGPT in a post on X (formerly known as Twitter). These strings mention a new clinician mode. It is described as a ChatGPT Workspace offering, but no other details were mentioned.

Netizens have speculated that the mode refers to the medical and healthcare consultancy capability of ChatGPT. While some believe that OpenAI is developing a dedicated medical GPT for this mode, it is likely that if a clinician mode is being built, it will be powered by GPT-5 itself.

The belief comes from the fact that the AI giant highlighted health as a major focus point of GPT-5 when it was introduced in August. At the time, the company had said, “GPT‑5 is our best model yet for health-related questions, empowering users to be informed about and advocate for their health.”

OpenAI also mentioned that, unlike older models, GPT-5 “acts more like an active thought partner” and can proactively flag potential concerns. The model was also said to be capable enough to ask questions to refine its responses based on the user's context, knowledge level, and geography.

The clinician mode could be a dedicated space for these capabilities, allowing users to take supplementary medical consultations from the chatbot before or after they visit a doctor. Notably, the AI giant did highlight that ChatGPT does not replace a medical professional.

The second feature mentioned in the strings is labelled as “model speaks first prompt.” The tipster claimed that this feature is likely to be added to the Voice Mode. Its description mentions, "Greet the user in a friendly way. The user's locale is {locale}, use their language. Be brief - no more than a few words."

As a note, just because features are mentioned in the code, it is no confirmation that it is being developed by the company. Developers often forget to remove code for discarded projects and add projects that are still in the pipeline. Often, these features never make it to the public. Unless OpenAI confirms these two capabilities, we would recommend taking this information with a pinch of salt.