OpenAI Presence Launches for Enterprise AI Agents
OpenAI has introduced OpenAI Presence, a product designed to deploy AI agents for customer and internal workflows, available as of July 22, 2026. This solution enables enterprises to utilize AI agents that can answer questions, resolve issues, and take approved actions while ensuring that human oversight is maintained when necessary.
Operational Framework of Presence
Each deployment of OpenAI Presence begins with a clearly defined task, such as resolving billing issues, supporting insurance claims, or managing employee IT service requests. The AI agent is granted only the knowledge and system access required for its specific role. Companies establish the policies that dictate what actions the agent can perform, when it requires approval, and when a human should intervene.
Testing and Quality Assurance
Prior to deployment, teams can rigorously test the agents against common requests, edge cases, and higher-risk scenarios. This is achieved through simulations and evaluators that assess whether the agent achieves the correct outcomes, adheres to company policies, utilizes tools appropriately, and escalates issues as needed. Guardrails are in place to intervene if interactions exceed the defined company boundaries.
Continuous Improvement Process
OpenAI has implemented a Codex-powered improvement loop that analyzes production sessions and escalations, suggesting updates through a Presence plugin. Teams can test these proposed changes against the live version before approving a controlled rollout. During initial testing, this improvement loop reportedly reduced human handoffs by 15 percentage points within just 10 days.
Proven Deployments and Current Use Cases
OpenAI Presence has been developed through extensive collaboration with its Research team and has been shaped by insights gained from years of enterprise deployments. Currently, it powers OpenAI’s English-language phone support channel at 1-888-GPT-0090, effectively handling open-ended requests, verifying callers, and utilizing account context to take approved actions. The system has met or exceeded benchmarks for human support, resolving 75% of inbound issues without human assistance.
Leading enterprises are also leveraging the same foundational technology. For instance, BBVA is exploring AI-driven voice banking solutions in Mexico, SoftBank is testing natural Japanese-language conversations, and IAG is evaluating support mechanisms for high-demand events.
Availability and Deployment
OpenAI Presence is currently available to eligible enterprise customers through a limited general availability program. Deployments are managed by OpenAI Forward Deployed Engineers (FDEs) in collaboration with select global systems integrators. It is important to note that Presence is not available as a self-serve product; organizations interested in exploring its capabilities should contact their OpenAI account team for eligibility details.
Original source: OpenAI News