Zoom broadens its developer platform with new AI Services, MCP expansion, and a unified platform studio
New voice, language, and intelligence APIs join expanded MCP support across Zoom products, with new integrations bringing Zoom capabilities into leading AI assistants, including Microsoft 365 Copilot
As businesses adopt more AI assistants and agents, developers face a growing challenge: connecting those tools to the information and capabilities scattered across enterprise applications. Building useful AI experiences often requires integrating multiple services and systems, which adds complexity and slows development.
Today at Zoomtopia, Zoom announced an expansion of its developer platform to address that challenge. New Zoom AI Services APIs give developers access to more of the AI technology behind Zoom’s products, expanded Model Context Protocol (MCP) support brings Zoom context and capabilities into external AI tools, and Platform Studio provides a single place to start building.
The approach reflects a broader shift in how Zoom is building for AI. Zoom will continue to create its own AI experiences while giving customers and developers the flexibility to use Zoom’s technology in the applications, models, and AI tools they choose.
“Businesses aren’t going to rely on a single AI assistant or model for everything they do,” said Brendan Ittelson, chief ecosystem officer at Zoom. “We believe the most valuable platforms will be the ones that work across that ecosystem, not just within their own applications. That’s why we’re opening more of Zoom’s technology to developers, so they can build on it and bring it into the experiences their customers choose.”
Build with the AI technology behind Zoom
Zoom AI Services gives developers direct access to speech, language, and voice technology based on the infrastructure Zoom uses for its own products. The APIs allow developers to add capabilities like transcription, translation, summarization, and voice to their applications without needing to build and operate the underlying AI infrastructure themselves.
At Zoomtopia, Zoom AI Services is expanding with 15 new services, adding conversational voice, real-time interpretation, advanced transcription, speech summarization, and customizable structured outputs.
Developers can now use the new Voice Agent APIs to combine speech recognition, LLM reasoning, and speech synthesis in a managed pipeline. They can choose an integrated model or bring their own LLM and connect it to Zoom’s speech-to-text and text-to-speech technology. New text-to-speech APIs support both synchronous and live-streaming applications, while the Interpreter API combines transcription, translation, and speech synthesis for near-real-time speech-to-speech interpretation.
Zoom is also adding ways to extract more useful information from spoken content. The new Speech Summarization APIs can turn audio directly into structured summaries and insights, with options to customize outputs for particular workflows. Expanded capabilities with Scribe API include automatic language detection, text polishing, time-synchronized captions, and new 3B Pro High-Accuracy models for batch, fast, and live transcription.
The new Decisions API lets developers evaluate text, images, and other inputs against predefined criteria and return structured results that applications can act on. Developers can use it to classify information, score responses, or determine whether specific conditions are met, helping automate tasks such as routing customer requests or flagging content for review.
The new Zoom AI Services APIs are now available for developers.
Extend context beyond the Zoom interface
Zoom is taking a headless approach to its platform, making its intelligence and capabilities available in the AI environments where people already work. Through Model Context Protocol (MCP), developers can bring Zoom context and supported actions into other applications and agents without requiring users to open Zoom.
Zoom is expanding It’s off-the-shelf MCP integrations with leading AI assistants. Alongside existing support for Anthropic Claude, OpenAI Codex, and Meta Muse, Zoom now supports Microsoft 365 Copilot, with Google Gemini support coming soon. These integrations allow users to work with supported Zoom capabilities and meeting context alongside other connected information, directly from their preferred AI assistant.
Zoom has also expanded its MCP support beyond meetings to more products across its portfolio, including Zoom Rooms and Workspace Reservation. Additional MCP servers are planned for Zoom Phone, Scheduler and Mail, Calendar, and Marketplace Build, giving developers more ways to connect AI agents to Zoom capabilities.
Developers can also use Search and Ask APIs to query Zoom’s unified context layer from their own applications and MCP-compatible agents. Zoom can retrieve relevant information across indexed sources and return that context to the application, reducing the need for developers to build separate search and retrieval workflows for each source. (Available now.)
A single place to start building
As Zoom makes more of its platform available to developers, Platform Studio provides a common entry point for building with it. Platform Studio brings together Zoom AI Services, Video SDK, and Co-Browse SDK in a single developer portal, giving developers an all-in-one location to access tools for building AI-powered communications experiences. (Available now.)
To learn more about Zoom’s developer tools and ecosystem, visit the Zoom website. For a look at Zoom’s AI Services API’s, visit Zoom.ai.
Related News
Zoom Media Kit
Explore our Media Kit for essential information and resources to help you out.
Learn more