Microsoft has begun scaling back its commitments with CoreWeave, a cloud provider specializing in AI workloads, just as the company prepares for a planned $4 billion IPO, according to the Financial Times.
The decision, attributed to delivery issues and missed deadlines, raises questions about CoreWeave’s financial outlook as it heavily depends on Microsoft, which accounted for 62% of its total revenue in 2024.
In a statement shared with WinBuzzer, CoreWeave refuted the claims made by the Financial Times, saying: “We pride ourselves in our client partnerships and there have been no contract cancellations or walking away from commitments. Any claim to the contrary is false and misleading.”
The move is part of a broader realignment in Microsoft’s AI cloud strategy, as the company prioritizes in-house infrastructure and chip development.
CoreWeave became a key partner for Microsoft after securing a multi-billion-dollar deal with the tech giant in June 2023. That agreement was driven by the need for large-scale AI infrastructure, particularly for OpenAI’s ChatGPT.
CoreWeave’s cloud platform, powered by Nvidia GPUs, allowed Microsoft to quickly scale its AI services. However, Microsoft’s latest pullback signals a shift in its reliance on third-party cloud providers.
CoreWeave’s IPO Faces New Risks
CoreWeave is seeking to raise $4 billion through its IPO, but Microsoft’s reduced commitments add uncertainty to its valuation.
The company reported $1.9 billion in revenue last year, but also recorded $863 million in net losses. To support its expansion, it raised $14.5 billion in debt and equity, including a $5 billion loan from Blackstone.
The company’s IPO filing explicitly warns about its dependency on Microsoft, stating that “any negative changes in demand from Microsoft, in Microsoft’s ability or willingness to perform under its contracts with us, in laws or regulations applicable to Microsoft or the regions in which it operates, or in our broader strategic relationship with Microsoft would adversely affect our business, operating results, financial condition, and future prospects.”
With Microsoft reducing its reliance on CoreWeave, the cloud provider now faces the challenge of proving to investors that it can sustain growth without its largest client.
Microsoft’s AI Cloud Strategy Is Evolving
Microsoft’s alledged withdrawal from parts of its CoreWeave agreements aligns with a broader trend of increasing control over its AI infrastructure.
The company recently ended its exclusive cloud deal with OpenAI, giving the AI research firm more flexibility in choosing compute providers, with OpenAI shifting towards SoftBank with new agreements.
Additionally, Microsoft has reportedly pulled out of two planned data center leasing agreements, according to analysts at TD Cowen, indicating a shift toward a more self-sufficient cloud strategy.
Despite reducing its reliance on third-party providers, Microsoft continues to invest heavily in AI, with around $80 billion allocated for AI-related cloud infrastructure in 2025. However, the company’s increased focus on in-house chip development suggests that CoreWeave’s role in its ecosystem may diminish over time.
Nvidia’s Role and GPU Supply Chain Challenges
CoreWeave’s success has been closely tied to Nvidia, which holds a 5% stake in the company. The cloud provider built its business around reselling access to Nvidia’s H100 GPUs, which power AI model training.
However, CoreWeave has recently faced challenges related to Blackwell GPUs, Nvidia’s next-generation AI chips.
Supply chain issues have delayed the rollout of these GPUs, impacting CoreWeave’s ability to meet demand. In its IPO filings, CoreWeave cited the risk of “asymmetry and delays” in its supply chain due to its reliance on Nvidia.
Nvidia CEO Jensen Huang also admitted in October that Blackwell chips had “design flaws” that contributed to shipment delays.
Microsoft, which has historically relied on Nvidia for AI hardware, appears to be diversifying its chip strategy. The company is investing in in-house chip development through its collaboration with AMD. This move could reduce Microsoft’s need for Nvidia-powered cloud providers like CoreWeave in the long run.
In addition to developing its own chips, Microsoft has been strengthening its Azure AI cloud platform to compete directly with AWS and Google Cloud. The move away from CoreWeave reflects a strategy of consolidating resources and prioritizing internal AI investments rather than outsourcing computing power to smaller providers.
Other major cloud companies are also reshaping their AI infrastructure strategies. Google and Amazon have been integrating their own AI-optimized hardware into their cloud offerings, reducing reliance on third-party suppliers. As Microsoft follows a similar path, the role of specialized cloud providers like CoreWeave may diminish unless they can secure a more diverse client base.
CoreWeave’s Business Model Faces Growing Pressure
CoreWeave’s rapid expansion has been fueled by its ability to deliver high-performance GPU compute power at a time when demand for AI infrastructure is surging. However, its reliance on a small number of customers and a single hardware supplier now puts it in a vulnerable position.
Microsoft’s decision to reduce its commitments means CoreWeave must demonstrate that it can attract and retain other major clients.
Unlike Amazon Web Services and Google Cloud, CoreWeave does not operate a broad-based cloud computing business. Instead, it focuses exclusively on GPU-based infrastructure, targeting AI firms that require large-scale computing resources.
While this specialization helped it secure Microsoft’s business, it also limits flexibility when faced with supply chain disruptions or changes in customer demand.
In its IPO filing, CoreWeave acknowledged the risks of its concentrated business model, stating that “any delay or shortage in GPU supply could materially impact our ability to deliver services to customers and fulfill contractual obligations.”
The delays in Nvidia’s Blackwell GPUs have added further strain to its operations. If CoreWeave is unable to secure enough alternative customers or adjust its pricing model, it could struggle to meet revenue targets.
CoreWeave’s Future Amid Increasing Market Pressures
CoreWeave’s ability to thrive in the evolving AI cloud market depends on how effectively it can adapt to the new competitive landscape. While it remains one of the largest independent AI cloud providers, Microsoft’s retreat forces it to reconsider its long-term strategy.
The company’s future will likely hinge on its ability to diversify its customer base, improve cost efficiency, and mitigate risks associated with relying on a single hardware supplier. As Microsoft, AWS, and Google continue investing in proprietary AI infrastructure, CoreWeave will need to prove that its model remains viable in a rapidly changing market.
