OpenAI launches Presence to bring AI agents into customer support and enterprise workflows

OpenAI has unveiled Presence, a new enterprise platform designed to help businesses deploy and manage AI agents across customer-facing and internal workflows. The offering is aimed at eligible enterprise customers that want AI agents capable of answering questions, accessing company systems, executing approved actions and escalating complex tasks to human employees, all while operating under company-defined policies, permissions and evaluation standards.

Presence is currently available through a limited general availability program. Deployments are led by OpenAI's Forward Deployed Engineers and select global systems integrators, meaning the product is not available on a self-service basis.

For users using non-OpenAI models

For enterprises that rely on non-OpenAI models, including increasingly popular open-weight models such as China's GLM-5.2 and Kimi K3, an OpenAI spokesperson clarified that "Presence uses OpenAI models for the core agent, while allowing customers to connect third-party models and services through APIs for guardrails, tools and other parts of their workflows."

More from Tech

The platform is designed to remain reliable as business rules, customer requirements and operating conditions evolve. Presence bundles together enterprise policies, system integrations, evaluations, guardrails and update processes required to deploy production-ready AI agents across business environments.

For organizations interested in adopting AI agents but unsure how to combine OpenAI's models, APIs, internal systems, security controls and evaluation tools into a production-ready solution, Presence is intended to simplify the process. Rather than building the underlying infrastructure themselves, enterprises work with OpenAI and its deployment partners to integrate AI agents into existing workflows.

Presence extended capabilities

At launch, Presence supports real-time voice and chat experiences, according to OpenAI's announcement. While the company's broader vision includes voice, chat, email and other communication channels, OpenAI has not confirmed that email support is available at launch.

Some of Presence's core capabilities include answering customer and employee questions in natural language, securely connecting with company systems and knowledge bases, and carrying out approved business actions. The platform can automate customer support by resolving routine inquiries, retrieving account information, processing approved requests and seamlessly handing conversations over to human agents when required. Internally, employees can use AI agents to access company policies, IT resources, HR information and organizational knowledge, while connected enterprise systems enable workflow automation that reduces manual effort.

Safety and Guardrails

A key design principle behind Presence is maintaining human oversight. Rather than replacing employees, the platform is built to recognize situations that require human judgment, automatically escalating complex, sensitive or exceptional cases to human operators. OpenAI says this approach allows businesses to improve productivity and customer service while retaining human control over critical decisions.

The launch reflects the industry's broader shift from AI assistants that simply answer questions to AI agents capable of carrying out meaningful work across enterprise operations.

The announcement also comes amid renewed scrutiny over AI safety after OpenAI and Hugging Face recently disclosed a security incident in which OpenAI's frontier models, during internal testing in the ExploitGym evaluation framework, reportedly exploited a previously unknown vulnerability, gained internet access and interacted with Hugging Face's systems while pursuing a benign benchmark-related task. Although the exercise took place in a controlled research environment, the incident raised fresh questions around AI sandboxing, permission controls, monitoring and enterprise security.