Microsoft launches new enterprise AI business with $2.5B investment

Tech giant Microsoft has launched Microsoft Frontier Company, a new business backed by a $2.5 billion investment that aims to help organisations deploy AI more effectively by combining AI engineering with industry expertise.

The move highlights a growing shift among businesses from experimenting with AI to demanding measurable returns while protecting their data and intellectual property.

The new business will embed 6,000 industry specialists and AI engineers within customer organisations to co-design, deploy and continuously improve AI systems. Rather than focusing solely on developing AI models, Microsoft said the initiative is designed to help businesses integrate AI into everyday operations and generate tangible business outcomes.

Announcing the initiative, Judson Althoff, chief executive officer of Microsoft Commercial Business, noted customers have moved beyond the early stages of AI adoption.

The announcement shows how enterprise AI is evolving. While the first wave of adoption centred on generative AI tools such as chatbots and assistants, organisations are increasingly looking to embed AI into business processes, decision-making and industry-specific workflows. Achieving that often requires combining AI models with a company’s own data, expertise and operational knowledge, alongside strong governance and security controls.

According to Althoff, companies need what he described as an “intelligence platform” where their proprietary data, expertise, workflows and decision-making processes continue to build value over time. Alongside this, businesses also require a trusted platform to observe, govern, manage and secure AI systems while monitoring costs and returns on investment.

Microsoft said Frontier Company combines AI engineering, industry expertise and organisational change management to create a continuous cycle of improvement rather than treating AI deployment as a one-time implementation.

Early projects illustrate that approach. Microsoft noted its engineers worked with the London Stock Exchange Group to integrate AI into LSEG Workspace, enabling finance professionals to ask complex questions across structured and unstructured financial information. The underlying system is continually refined using customer feedback and real-world usage to improve performance. The company also highlighted work with Land O’Lakes, Unilever and Novo Nordisk as examples of organisations already using the approach.

Protecting customer data and intellectual property is a central theme of the new business.

Althoff said, “A customer's IQ is protected. Their data, their IP, their competitive advantage, none of it is used to train models in ways that commoditize what differentiates them in their industry.”

Microsoft noted customers will be able to choose from multiple AI models, including those from OpenAI, Anthropic, Microsoft AI, open-source projects and specialised industry models, instead of being tied to a single provider. The company believes that this model-diverse approach allows organisations to select the most suitable AI system for different use cases while retaining control over their own data and knowledge.

The launch comes as competition in enterprise AI services intensifies. This week, Amazon Web Services announced a $1 billion investment in its own forward-deployed engineering organisation to help customers build AI applications. Companies such as Palantir have also expanded engineering-led AI consulting as businesses increasingly seek practical AI implementation rather than standalone software.

At the same time, governments in Europe, the United States and elsewhere are introducing AI governance frameworks that place greater emphasis on transparency, security and responsible deployment.

Microsoft has appointed Rodrigo Kede Lima as president of Microsoft Frontier Company to lead the new business.