OpenAI Just Released Its First Product That Isn't About the Model
The headline about OpenAI Presence is that it helps enterprises deploy AI agents. That isn't the story worth reading. The story is that Presence is the first OpenAI product where the model is deliberately invisible.
Launched July 22, Presence bundles policy controls, escalation rules, pre-deployment simulations, and a Codex-powered continuous improvement loop into a single enterprise framework for voice and chat AI agents. BBVA Mexico and SoftBank are the named first customers. Target use cases span customer support, outbound sales, HR, IT services, and procurement.
The deployment model is the tell. Presence is not self-serve. Every rollout is led by OpenAI Forward Deployed Engineers — the same subsidiary formed in May 2026 with $4 billion in backing and a $10 billion valuation. You don't buy Presence on a self-service portal. You get an FDE assigned to your deployment.
That's not a go-to-market quirk. It's a strategic declaration. OpenAI is telling the market that the model layer is solved — that what enterprises need now is not a better AI, but a trusted deployment partner who understands their workflows, their compliance requirements, and their escalation logic. Palantir used this exact playbook for a decade and built a $30+ billion company from it. OpenAI is a year into the same move.
The competitive frame just changed. OpenAI isn't competing with Anthropic and Google for frontier model contracts anymore. With Presence, it's competing directly with Salesforce Service Cloud, ServiceNow, and Zendesk for the enterprise workflow budget. Those companies sell software licenses. OpenAI is offering to own the outcome.
For investors building at the application layer, the implication is precise: the segments where Presence cannot profitably deploy — smaller markets, highly regulated contexts, vertical-specific workflows in LatAm, emerging market languages — represent the most defensible opportunity. OpenAI will take the Fortune 500 first. The rest is available.
| Metric | Value |
|---|---|
| Launch date | July 22, 2026 |
| OpenAI Deployment Company valuation | $10 billion |
| Deployment Company backing | $4 billion |
| First named enterprise customers | BBVA Mexico, SoftBank |
| Go-to-market model | FDE-led, not self-serve |
Frequently asked questions
What is OpenAI Presence?
OpenAI Presence is an enterprise AI agent deployment platform launched July 22, 2026. It bundles policy controls, escalation rules, pre-deployment simulations, and a Codex-powered behavioral improvement loop into a single framework for deploying voice and chat AI agents across customer support, sales, HR, IT, and procurement workflows.
Which companies are using OpenAI Presence?
BBVA Mexico is using Presence for customer interactions, and SoftBank has deployed Japanese-language agents through the platform. Every deployment is led by OpenAI Forward Deployed Engineers or select global systems integrators — it is not available self-serve.
How is OpenAI Presence different from ChatGPT Enterprise?
Presence is not self-serve — every deployment requires OpenAI Forward Deployed Engineers and is structured as a professional services engagement. ChatGPT Enterprise is a direct-access SaaS product. Presence competes with enterprise workflow software like Salesforce Service Cloud and ServiceNow, not other AI subscriptions.