OpenAI Introduces GPT Live for Real-Time Voice and Video Interaction

What OpenAI announced

OpenAI announced GPT Live, a new capability that lets ChatGPT see, hear, and respond in real time using voice and video. The feature is positioned as a step toward more natural, multimodal interaction with AI assistants.

How GPT Live works

According to the announcement, GPT Live enables continuous, real-time conversation with ChatGPT. Users can speak to the assistant, receive spoken responses, and share their camera feed so the model can interpret visual context on the fly. The system is designed to handle interruptions and follow-up questions without losing the thread of the conversation.

Core capabilities described

• Real-time voice conversation with natural turn-taking.
• Video input that lets ChatGPT analyze what the camera sees in the moment.
• Visual and on-screen awareness that can interpret the user’s environment.
• Integration within the ChatGPT experience, as described in the launch post.

Why this matters for multimodal AI

Voice and vision have been two of the most active areas of AI development. By combining them in a single real-time interface, GPT Live reflects a broader industry move toward assistants that perceive and respond to the world the way people do, rather than waiting for typed prompts.

For users, the practical impact is more conversational, less tool-like interaction with AI. For developers and businesses tracking the space, it signals that multimodal real-time interaction is becoming a baseline expectation rather than a research demo.

Availability and next steps

The launch post indicates that GPT Live is rolling out within ChatGPT. Specific rollout details, pricing tiers, and platform availability should be confirmed directly with OpenAI, since the source announcement could not be retrieved in full at the time of writing.