A Dedicated Arm for Enterprise AI Deployment
OpenAI has created the OpenAI Deployment Company as a majority-owned subsidiary focused on enterprise AI deployment at scale. Backed by more than USD 4 billion (approx. RM18.4 billion) in initial investment, the unit adds a services and engineering layer atop OpenAI’s models and APIs, targeting large organisations that want AI embedded in day-to-day operations. The strategy is to move beyond simple access to models toward full enterprise AI scaling—diagnosing where AI can add value, then designing and deploying robust OpenAI production systems. By positioning this as a standalone business division with its own operating model, OpenAI can specialise in AI pilot to production transitions while staying tightly linked to its core research, product, and internal deployment teams. The result is a direct bridge between frontier AI capabilities and the complex infrastructures that run modern enterprises.
Tomoro Acquisition: Embedding Forward Deployed Engineers
A central pillar of the new strategy is OpenAI’s agreement to acquire Tomoro, an applied AI consulting and engineering firm. Once regulatory approvals are complete, the deal is expected to bring about 150 Forward Deployed Engineers and deployment specialists into the Deployment Company. These teams are designed to embed within customer organisations, re-architecting critical workflows and infrastructure using frontier AI. Tomoro’s track record includes running real-time, large-scale AI systems for companies such as Tesco, Virgin Atlantic, and Supercell—contexts where governance, reliability, and integration are non-negotiable. This deep operational experience is crucial for shifting enterprise AI from isolated experiments to stable production systems. Instead of one-off proofs-of-concept, customers gain engineers who can tie OpenAI models directly to their private data, tools, and software, ensuring that enterprise AI deployment is grounded in real operational requirements rather than lab conditions.
From AI Pilots to Production Systems
The Deployment Company is explicitly designed to help organisations move from AI pilot to production. A typical engagement begins with a diagnostic phase, where engineers and business leaders identify the workflows most likely to benefit from AI. Rather than scattering experiments across departments, they select a small number of priority processes and design production systems around them. These systems connect OpenAI models to internal data, tools, controls, and processes, so AI becomes part of routine work instead of a standalone demo. This approach acknowledges that the bottleneck in enterprise AI scaling is no longer model access but integration into existing systems and governance frameworks. By linking customers directly to OpenAI’s ongoing model development, the unit aims to ensure that deployed solutions can evolve with new capabilities, avoiding the common trap of static pilots that never graduate into mission-critical OpenAI production systems.
Competing with Consultants and System Integrators
OpenAI’s move positions it squarely in competition—and collaboration—with traditional consulting and systems integration firms. The Deployment Company launches with backing from 19 investment firms, consultancies, and integrators, including TPG as lead partner alongside Advent, Bain Capital, Brookfield, and others. Consulting names like McKinsey & Company, Bain & Company, and Capgemini are also involved, bringing change-management expertise across thousands of client organisations. Collectively, these partners sponsor more than 2,000 businesses, giving OpenAI a broad pipeline of enterprise AI deployment opportunities. This structure reflects a wider market shift: software and model providers are pairing their technology with hands-on implementation and managed services. For large enterprises, the value lies in integrated teams that can work with executives, technology leaders, and frontline staff to redesign operations around AI, turning generic capabilities into tailored, measurable production outcomes.