Microsoft’s AI Data Centre Lease Cancellations: A Sign of Shifting Expectations?
Reports are circulating that Microsoft has pulled out of lease agreements for building data center capacity in the United States. This has sparked speculation about the tech giant’s expectations for artificial intelligence (AI) demand in a post-DeepSeek era. Microsoft has made significant investments in AI, and this move has raised eyebrows.
Satya Nadella, CEO of Microsoft.
Analysts at TD Cowen discovered that Microsoft canceled leases with private contractors for “a couple of hundred megawatts” of data center capacity in the US, according to a Bloomberg report. TD Cowen analysts also reportedly inquired with supply chain providers to uncover signs that Microsoft let agreements for data centers with more than a gigawatt of capacity expire. Additionally, Microsoft has withdrawn from multiple deals for building out data centers with 100 megawatts capacity each, the report states.
Why the Sudden Change?
Microsoft is one of the world’s largest data center owners and operators. The company has invested billions in securing AI computing capacity. While the reason for Microsoft’s purported move is not clear, TD Cowen suggested that the company cited “used facility” and “power delays” as reasons for terminating the data center leases. The analyst report further suggested that Microsoft may be reallocating some of its AI data center investments from overseas countries to the US.
“While we have yet to get the level of colour via our channel checks that we would like into why this is occurring, our initial reaction is that this is tied to Microsoft potentially being in an oversupply position,” TD Cowen analysts Michael Elias, Cooper Belanger, and Gregory Williams were quoted as saying by Bloomberg.
Impact of DeepSeek
The emergence of DeepSeek, a Chinese AI startup, may be affecting market dynamics. DeepSeek claims its R1 model rivals OpenAI’s and required fewer resources and less time to build than previously thought. This has led to fresh scrutiny of the massive AI spending plans of tech giants such as Google, Microsoft, and OpenAI.
In the wake of the DeepSeek-driven market changes, Microsoft CEO Satya Nadella argued that the Chinese AI startup’s influence might, counterintuitively, increase demand for advanced GPUs. Nvidia founder and CEO Jensen Huang has also stated that DeepSeek’s technological advancements will not negatively impact the chip giant’s business.
“I think the market responded to R1, as in, ‘Oh my gosh. AI is finished’. You know, it dropped out of the sky. We don’t need to do any computing anymore. It’s exactly the opposite. It’s [the] complete opposite,” Huang said in an interview.
Following the DeepSeek news, Nvidia’s share price declined significantly after a single-day drop. The company has lost a significant portion of its market value.
Microsoft’s Response
In response to the TD Cowen report, Microsoft reiterated its $80 billion commitment to AI data centers this fiscal year. “While we may strategically pace or adjust our infrastructure in some areas, we will continue to grow strongly in all regions. Our plans to spend over $80 billion on infrastructure this FY remains on track as we continue to grow at a record pace to meet customer demand,” a Microsoft spokesperson said.
Big tech companies, including Microsoft, Meta, Amazon, and Google, planned to spend a combined $320 billion in FY2025 to build data centers and purchase advanced graphics processing units (GPUs) to power their AI models. However, Microsoft’s purported lease withdrawals have re-ignited concerns about whether AI demand projections are overstated.
Critics have cautioned that the real-world applications of AI and paths for monetization remain unclear, despite tech giants doubling down on their AI spending commitments. Nadella stated in a podcast interview, “Us self-claiming some AGI milestone, that’s just nonsensical benchmark hacking to me. The real benchmark is the world growing at 10 per cent.” He also maintained his position that global computing needs for developing and running AI models will grow exponentially.
