Where Is the Next Stop for the AI Wave? Shift to Applications, Microsoft Poised to Capture the Monetization Prize

The artificial intelligence (AI) wave is driving global technology giants to make massive capital expenditures. Over the past two years, market attention has primarily focused on infrastructure areas such as data centers, chips, servers, and energy. However, as AI infrastructure gradually matures, market interest is beginning to shift from "who can provide computing power" to "who can generate real revenue through AI."

Goldman Sachs believes that the AI investment theme is undergoing a significant shift. The market is gradually transitioning from the infrastructure phase to the monetization phase. The growth model previously driven by continuous capital expenditure expansion from cloud service providers will now place greater emphasis on enterprise adoption, business models, and profitability.

The market's evaluation logic for AI infrastructure companies is also changing. In the past, investors focused on whether companies could benefit from increased capital spending by cloud giants. In this new phase, the ability of revenue, profit, and cash flow to mutually validate each other will become a key standard for assessing corporate value.

AI Demand Moving from Labs to Enterprises

The commercialization of AI is accelerating. In the past, there were concerns that cloud service providers were pouring large amounts of capital into AI model companies, creating a circular financing model of "passing money from one hand to the other," raising questions about the authenticity of demand for AI infrastructure spending.

However, enterprise demand is now emerging, and AI applications are gradually spreading from a few AI labs to traditional industries. Data shows that the growth in Microsoft's commercial remaining performance obligations primarily comes from non-lab customers, indicating that enterprise adoption of AI services is expanding.

This means that the value distribution within the AI supply chain is changing. The market is no longer just chasing companies that provide computing power, but is now seeking companies that can integrate AI technology into enterprise workflows, improve productivity, and establish revenue models.

Microsoft's Advantage in AI Monetization

In the AI monetization phase, Microsoft (MSFT-US) is seen as one of the most advantaged beneficiaries. Market analysis indicates that Microsoft and AWS are currently the only companies able to demonstrate proven AI business models, with Microsoft possessing multiple revenue streams such as Copilot and Foundry.

Microsoft's AI strategy forms a multi-layered revenue model. Azure provides foundational cloud computing power and GPU leasing services, the Foundry platform enables model invocation, fine-tuning, and deployment, and Copilot is directly embedded into enterprise application scenarios such as Office 365 and GitHub, generating revenue through a paid seat model.

This comprehensive strategy—from infrastructure to platform to end-user applications—enables Microsoft not only to benefit from AI capital expenditures but also to capture the broader business value created when enterprises adopt AI.

The Market Begins Seeking Answers on AI Profitability

As the AI industry enters its next phase, investor evaluation criteria will gradually shift from "how much was invested" to "how much return was generated." Over the past two years, cloud service providers' continuous increases in capital expenditure have driven growth across the AI supply chain. However, going forward, the market will place greater emphasis on whether companies can convert investments into actual revenue and profit.

Goldman Sachs judges that the AI theme is shifting from the "gold rush" phase of infrastructure investment to the phase of releasing application value. In this industrial transformation, companies equipped with cloud platforms, enterprise software, and AI application ecosystems are expected to become the next focal point of market attention—and Microsoft is a prime example.