The two critical components of any AI solution are intelligence and trust. This idea was first shared in November at our Ignite conference, and each conversation with customers only reinforces this belief. In my discussions, three main concerns consistently arise when companies consider adopting AI solutions:
- Will AI enhance my organization’s intelligence and the unique characteristics that differentiate us in our industry to help grow our business, or will it instead leverage our intelligence for its own gain by learning from key business operations and utilizing our intellectual property?
- Can we rely on AI solutions to provide a sustainable return on investment while operating within our governance and security standards?
- How can we achieve the necessary visibility, control, flexibility, and innovation in business models to effectively manage AI-related costs and maximize value?
I regularly advise customers to develop their own IQ on a platform built on diverse, open, and heterogeneous models at every level. AI models are becoming commodities, and reliance on a single model or harness should be avoided. Recently, Satya highlighted the risk of companies across sectors conceding value to a few dominant models. AI should foster growth by enhancing organizational intelligence from within.
Organizations should also have an observability platform that supports governance, management, security, and financial operations (FinOps) to ensure ROI with AI. This setup allows AI to be trusted in the environments it operates and gives businesses control over outcomes.
Intelligence and trust are integrated into products like Microsoft 365 Copilot, GitHub Copilot, and Copilot Studio, where model diversity ensures that costs and performance are aligned with tasks. Microsoft IQ optimizes workflows by efficiently directing context and reducing unnecessary computation. Agent 365 acts as the control plane for observing, governing, managing, and securing agents. We have designed a system to manage AI expenditure as a fundamental enterprise capability, delivered across various clouds and model providers without limiting customers to a single method.
Managing costs at scale
As agent usage increases, organizations need specific tools to manage costs:
- Model diversity: Relying on a single inferencing model or agentic loop does not compound an organization’s IQ. Both Microsoft 365 Copilot and GitHub Copilot are designed to be model-diverse, allowing customers to choose from various models like GPT-5.5 or Claude Opus 4.8, each excelling in different tasks and economic conditions.
- Your IQ: Agents struggle with raw data, and significant computing power is required to interpret structure and context before meaningful work can begin. The Microsoft IQ platform enhances your organizational IQ by converting raw data into actionable intelligence, creating a semantic understanding of operations across Microsoft 365 and business systems. This context is provided upfront, leading to faster execution, higher accuracy, and lower token usage.
- Financial operations: FinOps became crucial with the shift to the cloud and requires more attention as AI moves to usage-driven pricing models. With tools like Foundry and Agent 365, we help customers optimize their AI costs.
Frontier business models
Business models are evolving with AI to drive outcomes. The User Subscription License (USL) is foundational, offering a set of capabilities for a predictable monthly fee per user. Usage-based licensing has also emerged for agents handling multiple tasks, where costs align with the work performed.
Microsoft provides a unique combination of business model flexibility and integrated product experiences unmatched in the market. Microsoft 365 Copilot and GitHub Copilot employ both models, offering USL with added value and flexible consumption. We are announcing the global availability of Copilot Cowork, requiring the Microsoft 365 Copilot USL and based on usage.
Our model-diverse strategy allows customers to purchase capacity with the flexibility to choose the best model for each task, considering model strengths, economics, and innovations. Microsoft Agent Factory provides a unified consumption model spanning various tools, including Microsoft 365 Copilot and GitHub Copilot.
Our integrated product experiences bring AI into the workflow for knowledge workers and developers, managing capacity fluidly across both. The distinction between these roles is fading, with coding becoming a mainstream skill for knowledge workers, and chat and Cowork becoming vital for developers. With Microsoft 365 and GitHub, we offer leading tools for both roles, making it easy to manage capacity based on need.
Agent 365: The control plane
Organizations adopting agents from Microsoft or other providers, or building their own, need a control plane. Agent 365 provides IT and security leaders with a centralized platform to observe, govern, manage, and secure agents across the organization. It builds on trusted Microsoft technologies: Entra for identity, Defender for threat protection, Purview for data governance, and Intune for endpoint management. We are expanding Agent 365 to include cost management, enabling organizations to monitor agent spend alongside security and compliance.
The two crucial elements of any AI solution are Intelligence and Trust. At Microsoft, these principles guide the design of every layer of our AI platform. Microsoft IQ empowers organizations to leverage their unique intelligence, embedding AI into workflows to deliver swift, accurate, and reliable outcomes while safeguarding assets and intellectual property. Agent 365 offers a trust layer, ensuring all agents and AI artifacts are observed so organizations can confidently move from experimentation to enterprise impact. AI alone won’t transform businesses; the system managing it will. We’ve created this system for our clients, enabling intelligence to grow from within while maintaining control, visibility, and trust. Together, we can achieve significant business impact with AI across all roles and industries.
Judson Althoff is the CEO of Microsoft’s commercial business, overseeing product strategy, sales, services, support, marketing, operations, and revenue growth in over 120 subsidiaries worldwide.