OPENAI GIVES CHATGPT A VOICE THAT LISTENS AND TALKS AT THE SAME TIME AND NOTHING SOUNDS LIKE A ROBOT ANYMORE OpenAI released GPT-Live-1 on July 8 and within a day it became the default voice model for every paid ChatGPT user. The change is not a minor update. GPT-Live is built on full-duplex architecture, which means it can listen and speak simultaneously, the same way two people actually hold a conversation. The old system would wait for you to finish before responding. GPT-Live does not wait for anything. The model fills silences the way a person does. It says things like “mhmm” and “yeah” while you talk. It handles interruptions without losing track of what it was saying. When a question requires outside information, GPT-Live delegates to a background frontier model, pulls the answer, and continues the conversation without missing a beat. At launch that background model is GPT-5.5. Two versions are live. GPT-Live-1 goes to users on Go, Plus, and Pro plans. GPT-Live-1 mini covers the free tier. Both are rolling out globally across iOS, Android, and the web. This matters beyond the technical upgrade. OpenAI is betting that voice becomes the primary interface between people and AI agents. If GPT-Live-1 is any indication of where this goes, the keyboard may not hold its position as the default way humans and machines interact for much longer. The race to own how people talk to AI has moved from demos to daily use. Keywords: OpenAI GPT-Live-1, ChatGPT voice AI, full-duplex voice model, AI voice interface