GPT-Live

Overview

GPT-Live is OpenAI’s real-time, full-duplex voice experience and underlying voice model family designed for continuous conversation. It can listen and speak at the same time, respond with more natural conversational cues, and maintain a more fluid back-and-forth than traditional turn-based voice interfaces. OpenAI introduced GPT-Live as both a product experience in ChatGPT and a model offering with variants including GPT-Live-1 and GPT-Live-1 mini.

For AI Product Managers, GPT-Live matters because it signals a shift from basic voice commands to persistent, low-friction conversational interaction. Its rollout across ChatGPT Voice, desktop workflows, and business-focused plans suggests that voice is becoming a core interface layer for productivity, agent orchestration, and multimodal UX. PMs should view GPT-Live as both a user experience pattern and a platform capability for hands-free interaction, real-time assistance, and background delegation to more advanced reasoning models.

Key Developments

• 2026-07-09 — OpenAI introduced GPT-Live as a full-duplex voice model that can listen and speak simultaneously, use conversational fillers like “mhmm,” and hand off deeper search or reasoning tasks to GPT-5.5 in the background. Two versions, GPT-Live-1 and GPT-Live-1 mini, began rolling out to ChatGPT users globally, with an API sign-up also announced.
• 2026-07-24 — OpenAI launched ChatGPT Voice in its desktop app, powered by GPT-Live. The launch emphasized voice-based computer control and the ability to direct multiple agents in ChatGPT Work or Codex using spoken commands.
• 2026-07-28 — OpenAI expanded GPT-Live in ChatGPT Voice to Education, Business, and Enterprise plans globally, indicating broader organizational availability and stronger enterprise positioning.
• 2026-08-04 — OpenAI published an engineering-focused post on continuous voice interaction with GPT-Live, detailing technical work and UX improvements aimed at making ongoing voice conversations more reliable, natural, and useful in real-time settings.

Relevance to AI PMs

• Designing for continuous interaction: GPT-Live shows that voice products are moving beyond push-to-talk and rigid turn-taking. PMs should rethink conversational flows, interruption handling, latency expectations, and how products support ongoing dialogue without forcing users to restart context.
• Building multimodel experiences: GPT-Live’s delegation of deeper reasoning or search to GPT-5.5 suggests an orchestration pattern where a fast conversational layer fronts more capable background models. PMs can use this pattern to balance responsiveness, cost, and task quality.
• Expanding hands-free workflows: Its integration into ChatGPT desktop, ChatGPT Work, and Codex points to practical enterprise use cases such as agent coordination, voice-driven task execution, and accessibility-friendly interfaces. PMs should evaluate where voice can reduce friction in high-frequency workflows.

Related

• OpenAI — Creator of GPT-Live and the broader platform behind its rollout across ChatGPT products.
• ChatGPT — Primary product surface where GPT-Live was introduced and expanded to users.
• Advanced Voice Mode — A related voice interaction concept; GPT-Live appears to extend OpenAI’s push toward more natural, real-time spoken interfaces.
• GPT-5.5 — Referenced as the background reasoning/search model that GPT-Live can delegate more complex tasks to.
• ChatGPT Voice — Product experience powered by GPT-Live for real-time spoken interaction.
• ChatGPT Work — One of the agent-oriented workflows that users can reportedly direct by voice through GPT-Live-powered interfaces.
• Codex — Another workflow surface mentioned in connection with voice-directed multi-agent control.