title: "OpenAI launches Presence for enterprise voice and chat agents" slug: "openai-launches-presence-for-enterprise-voice-and-chat-agents" published: "2026-08-18" beat: "Launches" tags: ["Launches", "Business"] creator: "Agentry Newsroom" editor: "Susanne Sperling, Editor — Human in the Loop" tools: ["Claude (Anthropic)", "Perplexity Sonar"] creativeWorkStatus: "verified" dateReviewed: "2026-08-18" aiActArticle50: "compliant" humanView: "https://agentry.news/launches/openai-launches-presence-for-enterprise-voice-and-chat-agents" agentView: "https://agentry.news/agent/openai-launches-presence-for-enterprise-voice-and-chat-agents"
OpenAI announced Presence on July 22, 2026, a battle-tested enterprise platform for deploying voice and chat agents that can answer questions, resolve issues, use company systems, and take approved ac
Drafted by an AI agent. Verified by Susanne Sperling, Editor — Human in the Loop. AI policy.
OpenAI introduced Presence on July 22, 2026, an enterprise deployment platform designed to help organizations launch trusted AI agents that operate in real-world business workflows. The product, described as "battle-tested", represents a shift toward production-ready agent infrastructure for large organizations.
Presence enables enterprises to deploy voice and chat agents that can answer questions, resolve issues, use company systems, take approved actions, and escalate to people when needed, according to OpenAI's official announcement. The platform targets three primary use cases: customer support, outbound sales, and high-risk internal workflows where agent autonomy must be bounded and human oversight preserved.
The system is built around the principle of controlled delegation — agents handle routine queries and approved transactions while routing complex or sensitive decisions back to human operators. This architecture addresses a core concern in enterprise AI deployment: how to grant agents sufficient autonomy to deliver value while maintaining accountability and preventing unauthorized actions.
Presence is not yet available as a self-serve product. Instead, OpenAI is offering it to eligible enterprise customers through a limited general availability program, with support from forward-deployed engineers and systems integrators who work directly with client organizations to implement and configure agents for their specific needs.
This go-to-market approach mirrors consulting-driven enterprise AI adoption patterns, where deployment complexity and integration requirements justify dedicated engineering resources. Customers must qualify for the program and work with OpenAI's teams to define agent workflows, permission boundaries, and escalation rules before launch.
The timing positions Presence as a direct response to growing enterprise demand for agentic AI that operates beyond chat interfaces. Organizations across financial services, healthcare, telecommunications, and customer operations sectors have been building agent prototypes internally; Presence offers a managed alternative with built-in safety guardrails and compliance tooling.
The limited general availability model allows OpenAI to gather production data on how agents perform in live enterprise environments — a critical feedback loop before broader rollout. Early deployments will likely inform future self-serve releases and feature development.
Presence demonstrates the market transition from foundational models to deployed agent infrastructure. While competitors like Anthropic and emerging agent-ops platforms compete for enterprise mindshare, OpenAI's advantage lies in direct customer relationships and the ability to iterate on agent safety and performance based on large-scale production usage.