OpenAI Presence: Enterprise Agent Governance with High-Touch Deployment

OpenAI has launched Presence, a managed enterprise platform for deploying and governing AI agents across customer-facing and internal workflows. Instead of offering self-service access, Presence bundles guardrails, evaluations, and production monitoring with hands-on deployment by OpenAI's Forward Deployed Engineers (FDEs) and select system integrators. For AI builders evaluating enterprise agent infrastructure, Presence signals a shift from API-only access to a branded governance layer that controls agent behavior, permissions, and continuous improvement.

What happened

Presence packages company knowledge, standard operating procedures, approved actions, simulations, evaluation tools, guardrails, and escalation rules into a single platform. It starts with a defined job, such as resolving a billing issue or handling an IT request, and the agent receives only the information and system access required for that task. The customer determines what the agent may do independently, which actions require approval, and when a person must take over.

Before production, teams test agents against common requests, unusual edge cases, and higher-risk scenarios. Graders evaluate whether the agent reached the intended outcome, followed policy, used tools correctly, and escalated when required. Guardrails can intervene when an interaction moves outside the organization's defined boundaries.

After launch, Presence monitors production sessions, escalations, and quality signals. Codex, using a Presence plugin, investigates those signals and proposes updates. Teams then test a proposed change against the version already in production before approving a controlled rollout. This addresses the operational problem of agents becoming less reliable as policies, products, or user behavior change.

OpenAI says Presence already powers its English-language phone-support channel at 1-888-GPT-0090, resolving 75% of inbound issues without human assistance and reducing human handoffs by 15 percentage points over a 10-day period. These figures are company-reported and have not been independently verified.

Several large organizations are evaluating Presence. BBVA is exploring voice support for routine banking needs in Mexico. SoftBank is testing natural Japanese-language customer conversations. Australian insurer IAG is exploring support during high-demand periods such as severe weather and natural disasters.

Why AI builders should care

For AI builders, Presence represents a move beyond OpenAI's API business into a high-touch, governance-first deployment model. Instead of stitching together models, APIs, internal systems, security controls, and evaluation tools, enterprises can work with OpenAI's FDEs to put production-ready agents into existing workflows. This approach resembles Palantir's model of embedding engineers with customers, but Presence is more narrowly focused on agent behavior, approved actions, evaluations, escalation, and continuous improvement.

The launch comes as Anthropic also moves toward a services-led enterprise model through its consulting organization Ode. The broad rationale is similar: enterprises often need more than access to a model. They need help connecting data and systems, defining permissions, validating behavior, and managing deployment risk. Presence differs by packaging those requirements into a branded agent-governance product with a defined operational layer for policies, simulations, evaluations, approvals, and production updates.

For builders evaluating agent infrastructure, Presence introduces a new variable: the cost and dependency of high-touch deployment. It does not replace OpenAI's API business, but it offers an alternative for organizations that cannot or do not want to build their own governance layer.

Practical implications

Presence is available immediately through a limited general availability program. Deployments are led by OpenAI Forward Deployed Engineers and select global systems integrators. The product is not available on a self-service basis.

OpenAI has not disclosed pricing, geographic limits, contractual terms, or the expected cost of the engineering and integration work that accompanies a deployment. The company also has not said whether Presence can use models from providers other than OpenAI. These unknowns make total-cost-of-ownership assessments difficult for potential customers.

Presence currently supports real-time voice and chat experiences. OpenAI's outreach materials describe a broader ambition spanning voice, chat, email, and other channels, but email support has not been confirmed at launch.

Caveats

Several caveats are worth noting. The 75% issue resolution rate and 15-point handoff reduction are company-reported and have not been independently verified. Pricing, SLAs, and interoperability details remain undisclosed