Microsoft launches $2.5bn Frontier company to help businesses adopt AI

By Ochiabuto Nnaji

Microsoft has launched a new enterprise-focused business, Microsoft Frontier Company, backed by $2.5 billion in funding, to help large organisations identify, integrate and manage artificial intelligence (AI) technologies from multiple providers.

According to Reuters, the new venture will advise companies including Unilever and Novo Nordisk on selecting AI models best suited to their operations and combining them with proprietary business data to generate measurable returns on investment.

The initiative reflects a growing trend among enterprises to move beyond relying on a single AI vendor and instead deploy a mix of AI models tailored to different business functions.

Unlike conventional AI consulting arrangements, Microsoft said customers will retain ownership of the work and intellectual property generated through the engagements, addressing growing concerns among businesses over data sovereignty, confidentiality and competitive risks.

Speaking to Reuters, Chief Executive Officer of Microsoft Commercial Business, Judson Althoff, said the company’s experience developing its Microsoft Copilot platform shaped the new strategy.

“Three years ago, when we built Copilot, we made a mistake by binding it to OpenAI models only,” Althoff said, adding that enterprise customers increasingly value the flexibility to switch between AI models as technology advances.

According to him, businesses are less concerned about which AI model they use than about how effectively those models work with their own proprietary data and how easily they can adopt newer technologies as they emerge.

Microsoft Frontier Company will therefore integrate Microsoft’s AI capabilities alongside models from external providers, enabling customers to build customised AI systems that best suit their operational requirements.

The launch underscores a broader shift in the enterprise AI market, where organisations are increasingly adopting multiple AI models—including open-source systems—instead of depending exclusively on providers such as OpenAI or Anthropic.

While the multi-model approach offers greater flexibility and performance, it also increases implementation costs and complexity, often delaying returns on investment. Microsoft aims to simplify that process by acting as a neutral AI integration partner.

The strategy also reflects Microsoft’s evolving role within the AI ecosystem. Although the company is a major investor in OpenAI and recently expanded Copilot to support Anthropic’s AI models, Frontier Company signals a shift toward positioning Microsoft as a trusted platform capable of orchestrating technologies from multiple AI developers rather than promoting a single model.

Industry analysts say the approach aligns with growing concerns among large enterprises about becoming overly dependent on any one AI provider.

Chief Executive Officer of analyst firm Moor Insights & Strategy, Patrick Moorhead, noted that many corporations fear giving frontier AI companies too much insight into proprietary business processes could eventually enable those firms to compete directly with their customers, particularly in sectors such as software development and legal services.

The announcement comes as Microsoft continues to reshape its business around artificial intelligence. Earlier this week, the technology giant disclosed plans to eliminate thousands of jobs across its sales, consulting and Xbox gaming divisions as it reallocates resources toward AI development and infrastructure investments. Reports indicate the layoffs will affect fewer than 2.5 per cent of Microsoft’s global workforce of about 220,000 employees.