Azure’s quarterly revenue increased by 43%; Microsoft says the investments in AI are starting to yield financial returns.

As the market continues to ask when the investments in AI will yield returns, Microsoft has presented results that are closer to achieving commercial success. According to a report by Reuters on July 30, Microsoft’s Azure business saw its revenue increase by 43% on a year-on-year basis in the fiscal quarter that ended in June, exceeding analysts’ expectations of around 39.98%.

Microsoft also expects Azure to experience a 45% growth rate on a fixed exchange rate basis in the new fiscal quarter, which is higher than market expectations. The company’s overall quarterly revenue increased by 18% to $90 billion. After the release of these results, Microsoft’s stock price rose by more than 8% in after-hours trading, indicating that investors’ concerns regarding the company’s investment in AI infrastructure have eased.

What is even more noteworthy are the data related to product usage. The number of paid subscriptions to Microsoft 365 Copilot has exceeded 30 million; the figure reported in the previous quarter was 20 million, which is also higher than what analysts had predicted. Corporate clients are moving from piloting generative AI solutions to actual purchases of these services, thereby creating a clearer revenue stream from cloud computing resources, software subscriptions, and model services.

Microsoft’s CEO, Satya Nadella, said that the company is working on developing its own models and chips, which should result in an efficiency improvement of up to 40%. In the past, Microsoft relied heavily on OpenAI for its core models; now it is moving toward a multi-model architecture that allows both the company and its customers to choose different technologies based on cost, performance, and the type of task at hand.

However, rising returns do not mean that investment levels have reached their peak. Microsoft expects to spend around $50 billion on capital expenditures in the first quarter of fiscal year 2027, and it has disclosed that its commitments regarding data center rentals amount to $329.1 billion – commitments that have not yet been fulfilled. The company is turning the demand for AI solutions into cloud-based services, but it still has to bear long-term costs related to data center infrastructure, electricity, networking, and equipment upgrades.

These results send two signals for these industries: first, enterprise-level AI is moving from the stage of conceptual validation to large-scale adoption; second, model capabilities are only one aspect of business competition – platforms that possess cloud infrastructure, development tools, customer access points, and sales networks are better able to turn AI usage into sustainable revenue. Whether Microsoft can maintain the growth of Azure and improve its inference efficiency will be key to determining the return on investment in AI in this round.

Source:Reuters