Agentic coding goes hands free as OpenAI brings GPT-Live's full duplex voice control to Codex and ChatGPT on the desktop

OpenAI announced Thursday it is integrating its GPT-Live full-duplex voice model into the ChatGPT desktop application for macOS and Windows, bringing hands-free voice control to agentic coding systems Codex and ChatGPT Work. The company said the update allows software engineers to orchestrate multi-threaded coding jobs, review pull requests, and debug applications using natural voice commands. The integration relies on decoupling the real-time voice layer from underlying execution engines. While GPT-Live maintains fluid conversation, it passes heavy computational workloads to background reasoning models. On macOS, the desktop application incorporates "Appshots" and screen context features, allowing ChatGPT Voice to analyze the frontmost window alongside local files and codebase structures. Developers can initiate multiple concurrent task threads from a single spoken prompt, such as investigating a bug, reviewing a pull request, and generating unit tests simultaneously. OpenAI posted a promotional video showing Codex developer experience engineer Jason Liu and Codex technical staffer Guinness Chen speaking to the same ChatGPT desktop app session in the same room. Access is restricted to paid subscribers across Plus, Pro, Business, Enterprise, and Education plans. The company said tasks initiated via ChatGPT Voice consume standard usage allocations from existing Codex and ChatGPT Work plan quotas.