Microsoft is reportedly set to launch its own artificial intelligence (AI) chip in a bid to reduce its reliance on Nvidia. This move aligns with Microsoft’s ambitious plans in the AI space. According to sources, the chip will be similar to Nvidia’s graphics processing units (GPUs) and will be specifically designed to run data center servers for training and operating large language models, which are foundational to generative AI technology. The chip’s debut is expected to take place at Microsoft’s Ignite conference in November.
This development comes shortly after OpenAI, the firm behind ChatGPT, revealed that it is considering creating its own AI chips. OpenAI is also exploring options like collaborating more closely with chip manufacturers, including Nvidia, as well as diversifying its suppliers and potentially making an acquisition.
The demand for AI chips has surged with the rise of generative AI. Companies that venture into manufacturing their own chips will be able to control costs and address potential supply shortages. Meta Platforms, formerly known as Facebook, is already developing its own silicon chip, Meta Training and Inference Accelerator (MTIA), to support its AI ambitions. Amazon has quietly been working on its own chip since 2013 and has now accelerated its efforts in the generative AI race.
While Nvidia has become a leading player in the chip industry thanks to its powerful semiconductor offerings for generative AI, Oracle has pledged to spend billions on Nvidia chips to bolster its position in both generative AI and cloud computing. Additionally, SoftBank-owned Arm, a British semiconductor unit, recently went public, raising close to $4.87 billion in its initial public offering.
According to research firm Gartner, AI-specific chips are projected to generate 21% more revenue this year, reaching $53.4 billion. This growth is largely attributed to advancements in generative AI. Gartner forecasts that revenue will accelerate next year, surpassing $67.2 billion, and more than double to nearly $120 billion by 2027. In the consumer electronics market alone, the value of AI-enabled application processors used in devices is estimated to more than double each year and reach $1.2 billion by the end of 2021.
As Microsoft prepares to unveil its own AI chip, the tech industry is witnessing a trend towards companies manufacturing their own chips to meet the increasing demand for AI capabilities. This move will not only enable these companies to reduce costs but also strengthen their AI ambitions and potentially gain a competitive edge.