Microsoft puts $2.5 billion into a new unit aimed at deploying AI inside enterprise customers

Microsoft said it will commit $2.5 billion to launch a new business focused on helping enterprise customers understand and implement artificial intelligence, moving beyond packaged software to hands-on deployments inside client organizations. The initiative, described as Microsoft Frontier Co., will combine engineering, technical consulting, support and sales resources and place them alongside customers as those organizations build and operate AI systems.

Microsoft also said the new unit will draw on 6,000 employees who will be embedded with clients. The company’s approach is described as forward deployed engineering, a model in which engineers and customer-facing teams work directly within customer projects rather than waiting for demand to arrive through standard services. Microsoft said the unit will include existing forward deployed engineering teams, plus industry-focused technical consultants and support staff, and sales personnel with relevant domain experience.

CNBC reported that Microsoft plans to move engineers and salespeople into the new subsidiary to work closely with clients on AI engagements. The report also said the unit will carry a mix of functions, including deployment and ongoing operational work once AI systems go live, reflecting the company’s view that implementation and integration are where many enterprise AI efforts stall.

Microsoft Frontier Co. is positioned amid a competitive rush among major technology companies to build internal capacity for AI deployments. CNBC noted that Amazon has backed its own forward deployed engineering initiative with $1 billion, while AI labs such as Anthropic and OpenAI have established forward deployed engineering groups, partnering with private equity firms, banks and consulting partners. The pattern highlights how vendors are increasingly treating AI implementation services as a differentiator rather than leaving it to partners.

The launch comes as Microsoft continues to build the underlying infrastructure and tools used by enterprises experimenting with generative AI. Microsoft has invested tens of billions of dollars in data centers to support generative AI workloads and has released a portfolio of AI services. Still, Microsoft has faced mixed results in adoption across those products, including with Copilot and GitHub Copilot, according to the CNBC reporting.

A key figure in the new unit is Rodrigo Kede Lima, who CNBC said will serve as president of Microsoft Frontier Co. The appointment suggests Microsoft intends to operationalize AI delivery in a way that mirrors traditional field-sales and delivery models, but with specialized engineering leadership and industry coverage embedded in the client relationship.

What Microsoft did not spell out in the announcement and related reporting is how the $2.5 billion will be allocated across hiring, delivery centers, compensation, or partner programs, or whether customers will be charged separately for the embedded teams. The company’s disclosures also did not specify performance metrics, target industries, or timelines for scaling the unit across regions, leaving those details unclear.

For customers and competitors, the immediate question is whether Microsoft can convert AI interest into faster, more reliable deployment outcomes. If forward deployed engineering proves effective at moving organizations from pilots to production, it could reshape how enterprises procure AI, with vendors bundling implementation capacity more tightly into cloud and AI offerings. Microsoft’s next announcements on service scope, contracting terms and early customer outcomes will likely determine how quickly the market views Microsoft Frontier Co. as a durable capability rather than a one-off push.