Microsoft confirms its own chips development to support IA operations

The growing demand for artificial intelligence is redefining the global semiconductor market. Microsoft, One of the main actors in the cloud computing sector has intensified its commitment to the development of own chips, this while reliating in Nvidia and AMD For much of its current operations.

During a conversation in the Italian Tech Weekmoderated by CNBC, Kevin Scott, director of Technology of Microsoft, He explained the company’s vision against the creation of personalized semiconductors for artificial intelligence. According to the manager, the priority is to ensure “The best performance-price” In each chip, which has led to maintaining the flexibility of working with different suppliers.

The importance of Nvidia and the transition to your own chips

Scott said Nvidia has led the market due to its ability to offer the best cost and performance combination. However, he emphasized that Microsoft It does not rule out working with other manufacturers to guarantee sufficient capacity against the growing demand.

In parallel, the company has already begun to integrate its own hardware. In 2023 he presented the Azure Maia Ai Acceleratorspecifically designed for artificial intelligence work loads, together with the processor Cobalt CPU. Recent reports point to that Microsoft He works in a new generation of semiconductors that will strengthen their technological independence.

The Executive confirmed that the long -term goal is to progressively migrate towards greater use of own manufacturing chips in their data centers. “Absolutely”, Scott responded to being consulted on this strategy, highlighting that they currently use “Much Microsoft silicon.”

Innovation in refrigeration and infrastructure management

One of the biggest challenges in the industry is the dissipation of heat generated by chips. Last week, Microsoft presented a new cooling technology through “Microfluid”, Designed to reduce overheating and increase energy efficiency in your data centers.

Scott stressed that the company’s plan goes beyond hardware, covering network systems, refrigeration and the ability to adapt the infrastructure to different computer loads. The strategy reflects a comprehensive approach in which not only the design of chips matters, but also the way they are integrated into the global technological ecosystem.

Microsoft He is not alone in this bet. Companies like Google, Amazon and Goal They have also announced the development of custom chips, with the double objective of reducing the dependence of Nvidia and to optimize your data centers for specific artificial intelligence tasks.

This movement is part of a record year of investments in the sector. According to reports, the main technological ones will allocate more than USD $ 300,000 million In capital expenses in 2025, a large part of them aimed at strengthening their AI capacities.

Innovar pressure responds to the need to stay competitive in front of a market that has shot Chatgpt and other generative models. Companies seek not only save costs, but also build more resilient and scalable infrastructures.

The limitations of computing capacity persists

Despite these investments, the availability of computing capacity remains a critical problem. Scott said that describing the situation as a “Mass shortage” It does not completely reflect the magnitude of the challenge. From the irruption of Chatgpt, Demand has exceeded all initial projections.

The cto of Microsoft He admitted that even his most optimistic predictions were insufficient to anticipate explosive growth in the consumption of artificial intelligence resources. Although the company has opened multiple data centers and plans an even greater expansion in the coming years, the gap between supply and demand persists.

The global shortage of computing capacity has become a pressure point for the entire industry. While technological giants accelerate investments in their own infrastructure and chips, the panorama indicates that the balance between demand and supply could take several years to stabilize.

Written article with the help of an AI content editor, edited by Angel Di Matteo / Diariobitcoin

Original image of Diariobitcoin, created with artificial intelligence, for free use, licensed under public domain.

WARNING: Diariobitcoin offers informative and educational content on various topics, including cryptocurrencies, AI, technology and regulations. We do not provide financial advice. Cryptactive investments are high risk and may not be adequate for all. Investigate, consult an expert and verify the applicable legislation before investing. I could lose all its capital.

Subscribe to our newsletter