Microsoft has been seeking more GPUs due to the interest in ChatGPT and its reliance on Microsoft's Azure cloud for computations. The company has also started using OpenAI's models to improve its products, such as Outlook, Word, and Bing. To secure enough capacity, Microsoft has signed an agreement with Nvidia-backed CoreWeave, which rents out GPUs as a cloud service. The company also plans to increase its capital expenditures this quarter to pay for data centers, CPUs, networking hardware, and GPUs.
Key takeaways:
- Microsoft is emphasizing the importance of graphics processing units (GPUs) for its growing cloud business, adding language about GPUs to a risk factor for outages in its annual report.
- The company's demand for GPUs has increased due to the popularity of AI, particularly generative AI, and the success of OpenAI's ChatGPT chatbot, which relies on Microsoft's Azure cloud for computations.
- Microsoft has started looking outside its own data centers to secure enough capacity, signing an agreement with Nvidia-backed CoreWeave, which rents out GPUs as a cloud service.
- Microsoft is also accelerating the deployment of its own custom AI processor, and expects to increase its capital expenditures this quarter to pay for data centers, CPUs, networking hardware, and GPUs.