AI Models & Platforms

OpenAI Brings Plugins to Live Voice and Voice to Work in ChatGPT

mm
Published
Add Unite.AI to your preferred sources on Google

OpenAI on September 23, 2026, added plugin support to the Live voice experience in ChatGPT across web, iOS, and Android and brought Voice into ChatGPT Work on web and mobile, according to the company’s ChatGPT release notes. The update lets users invoke connected apps during a spoken conversation and start Work tasks by voice from a phone or a browser.

Plugins in Live Voice

The September 23 entry states that Live now supports plugins on web, iOS, and Android. During a Voice conversation, users can draw on the plugins and connected apps tied to their account and follow the written responses in the chat. Free and Go users can use Voice in Chat with the plugins their plan supports, and an account’s existing app connections, permissions, and usage limits carry over.

OpenAI’s ChatGPT Voice help article documents the mechanics. During a Live conversation, a user asks ChatGPT to use a plugin available to the account. The user may first need to connect the app and sign in, and existing app permissions and workspace restrictions still apply. When an action requires approval on web or mobile, Voice prompts an on-screen review with on-screen controls to approve or decline, and OpenAI states that approving by voice is not supported.

Voice in ChatGPT Work

The release notes state that Voice is now available in Work on web and mobile. By speaking, users can have Work draft documents, presentations, and spreadsheets, draw on connected apps, or operate in a browser. If a task is still running when the user ends the voice call, it can continue in text. Voice in Work requires both Voice and Work access on the account.

According to the help article, Voice in Work follows the plan’s Voice limits and billing, and tasks started by voice draw on normal Work usage just as typed tasks do. The article separately documents Voice in the ChatGPT desktop app on macOS and Windows, where users can start tasks, check progress, ask questions about their agents, and coordinate multiple agents through one conversation.

Voice Options, Usage Limits, and Stated Limitations

The help article lists three Voice options. Live runs on GPT-Live-1 or GPT-Live-1 mini depending on the user’s plan, and OpenAI describes it as able to listen and speak at the same time, use web search and memory, show visual results through supported widgets, and work with text and images in the same chat. Advanced is the previous real-time Voice experience and keeps supported mobile capabilities such as video and screen sharing, which Live does not support. Standard is a turn-by-turn option that transcribes the user’s speech before producing a response.

For Voice in Chat, Live usage is measured over a rolling 24-hour period. The article’s posted limits give Free users limited access to GPT-Live-1 mini, Go users up to 3 hours with GPT-Live-1 mini, Plus users up to 3 hours with GPT-Live-1, Pro subscribers at $100 per month up to 15 hours, and Pro subscribers at $200 per month unlimited GPT-Live-1 usage. Business Standard receives 3 hours with GPT-Live-1, with additional usage consuming 1.25 credits per minute, while Business Premium receives 15 hours. Enterprise, Edu, and Clinicians workspaces on credit-based pricing consume 1.25 credits per minute, and Enterprise on usage-based pricing is billed at $0.05 per minute.

OpenAI also states two current limitations for Live: it does not support video or screen sharing, and it cannot currently find or add files from a user’s ChatGPT Library, though a supported file may still be attached to the chat manually depending on the account.

Documented Rollout Path

The September 23 update extends a sequence recorded in the same release notes. On July 8, 2026, OpenAI introduced GPT-Live-1 in ChatGPT Voice for paid users and GPT-Live-1 mini for Free users, models the company said can listen and speak at the same time so that turn-taking and interruptions feel more natural, with web search, memory, and visual widgets across chatgpt.com and the iOS and Android apps. On July 9, 2026, the company introduced ChatGPT Work, an agent for longer, multi-step work that researches and analyzes information, operates across connected apps and files, and produces documents, spreadsheets, presentations, reports, and Sites, rolling out on web and mobile to paid plans except Free and Go. On July 23, 2026, ChatGPT Voice became available in Work and Codex in the desktop app.

On August 31, 2026, content from a Live voice conversation gained the ability to appear on an iPhone Lock Screen and, on supported iPhones, in the Dynamic Island through Live Activities. On September 9, 2026, OpenAI let Voice draw on GPT-5.6 or GPT-6 Astra when it needs to search or reason through harder questions, simplified GPT-Live daily usage limits by plan, ended the practice of switching Plus and Pro users to GPT-Live mini after they reached a Voice limit, and deprecated the separate Instant, Medium, and High Voice intelligence levels.

mm

Jonas Reeve is an AI-generated analyst at Unite.AI, focusing on cognitive AI, artificial general intelligence (AGI), and the theoretical foundations of machine intelligence. His work explores how learning, reasoning, memory, and abstraction emerge in both biological and artificial systems, drawing connections between modern AI architectures and long-standing questions in cognitive science and philosophy of mind.

With a conceptual and reflective approach, Jonas examines frameworks such as reasoning models, agentic systems, emergent cognition, and alignment theory, aiming to clarify what progress toward AGI actually means—and what it does not. Rather than chasing timelines or hype, he emphasizes first principles, conceptual rigor, and the limits of current models.

Articles authored by Jonas Reeve are AI-generated and reviewed by Unite.AI’s editorial team to ensure accuracy, clarity, and responsible discussion of advanced AI concepts.