OpenAI is changing how businesses buy its AI agents. The company announced Presence on July 22, a managed service that sells enterprise AI agents as projects rather than products. Each deployment comes with OpenAI's own Forward Deployed Engineers and selected global systems integrators.
How Presence Works: Projects, Not Products
Presence is not a self-serve product. OpenAI states plainly that it is not yet available for direct purchase. Instead, each engagement starts with a single job — resolving a billing dispute, handling an insurance claim, or clearing an employee IT service request. The agent is given only the knowledge and system access needed for that specific task.
Why This Matters for Enterprise Buyers
This is a departure for a business that has largely run on API keys and seat licences. Presence is sold as a project, meaning enterprises get a custom deployment led by OpenAI's own engineers. For companies that need AI agents to handle sensitive or complex workflows, this managed approach reduces risk and ensures the agent works within existing systems.
What OpenAI's Forward Deployed Engineers Do
OpenAI's Forward Deployed Engineers are the same team that handles high-stakes enterprise deployments. They work directly with clients to configure the agent, integrate it with existing systems, and ensure it performs the specific job. Selected global systems integrators also participate, providing additional support for large-scale rollouts.
Who Is Affected and What It Means
Enterprise IT leaders, customer service heads, and operations managers are the primary audience. For them, Presence offers a way to deploy AI agents without building in-house expertise. The service is currently in limited general availability, meaning only select clients can access it. OpenAI has not announced when or if a self-serve version will be available.
Official Response and Limited GA Programme
OpenAI announced Presence on July 22 through its official channels. The company stated that the service is delivered through a limited general availability programme and is not yet available as a self-serve product. Deployments are led by OpenAI's Forward Deployed Engineers and a set of selected global systems integrators.
How Presence Differs from OpenAI's Usual Model
OpenAI's typical business model relies on API keys and seat licences. Customers pay for access to models like GPT-4 and use them within their own applications. Presence flips this model: instead of selling access to a model, OpenAI sells a managed project that includes the agent, the engineers, and the integration work. This is closer to a consulting engagement than a software purchase.
Confirmed Facts vs What Remains Unclear
Confirmed: Presence is a managed enterprise AI agent service announced on July 22. It is sold as a project, not a product. Deployments are led by OpenAI's Forward Deployed Engineers and selected global systems integrators. It is currently in limited general availability with no self-serve option. Unclear: Pricing details, number of clients in the programme, timeline for wider availability, and whether a self-serve version will ever be offered.
Why OpenAI's Approach Matters for the AI Industry
OpenAI's move signals that enterprise AI deployment still requires human expertise. Many companies struggle to integrate AI agents into existing workflows. By offering a managed service with engineers attached, OpenAI addresses the adoption barrier head-on. This could set a precedent for how AI companies sell to large enterprises — not just as software, but as a service with human support.
Risks and Balanced View
Presence is not for every business. The limited GA programme means only select clients can access it. Pricing is unknown, and the managed model may be expensive compared to self-serve alternatives. Critics may argue that OpenAI is creating a dependency on its own engineers, making it harder for clients to switch providers. Additionally, the project-based model may not scale easily for companies with many use cases.
Wider Trend: AI Agents as Managed Services
OpenAI is not alone in this approach. Other AI companies are also moving toward managed deployments for enterprise clients. The trend reflects a growing recognition that AI agents need careful configuration, integration, and monitoring. Selling AI as a managed service reduces risk for buyers and creates recurring revenue for sellers.
Practical Guidance for Enterprise Leaders
If your company is considering AI agents for specific tasks like billing disputes, insurance claims, or IT service requests, Presence may be worth exploring. However, the limited GA programme means you may need to wait. In the meantime, evaluate your existing workflows and identify which single job would benefit most from an AI agent. Prepare your data and system access requirements so you can move quickly when the service becomes more widely available.
Future Outlook
OpenAI has not announced a timeline for wider availability of Presence. The limited GA programme will likely generate feedback that shapes the final product. If successful, Presence could become a template for how OpenAI sells to large enterprises. A self-serve version may eventually follow, but for now, the company is focused on managed deployments with human engineers attached.
Our Take
OpenAI Presence represents a pragmatic shift in how enterprise AI is sold. By attaching engineers to each deployment, OpenAI acknowledges that AI agents are not plug-and-play. They need careful configuration, integration, and oversight. For enterprises, this managed approach reduces risk and increases the likelihood of success. However, the limited availability and unknown pricing mean it is not yet a mass-market solution. The real test will come when OpenAI decides whether to scale Presence as a self-serve product or keep it as a premium managed service.
Frequently Asked Questions
What is OpenAI Presence?
OpenAI Presence is a managed enterprise AI agent service announced on July 22. It is sold as a project, not a product, with deployments led by OpenAI's Forward Deployed Engineers and selected global systems integrators.
How is Presence different from OpenAI's usual products?
Unlike API keys and seat licences, Presence is a managed service. Each engagement starts with a single job, and OpenAI's engineers handle the deployment and integration. It is not available as a self-serve product.
Who can use OpenAI Presence?
Presence is currently in limited general availability. Only select enterprise clients can access it. OpenAI has not announced when or if a self-serve version will be available.
What kind of jobs can Presence handle?
Each engagement starts with a single job, such as resolving a billing dispute, handling an insurance claim, or clearing an employee IT service request. The agent is given only the knowledge and system access needed for that specific task.
