The AI Revolution: CPU Makers Going Crazy

The image does not belong to buddystech.com

There is a lot of talk in the CPU manufacturing industry about how the rise of artificial intelligence (AI) is changing things. As generative AI apps like ChatGPT and Sora become very popular, the need for computing power is going through the roof². Companies that make CPUs are working hard to keep up with the demand, which is making them crazy.

The Rise of AI

As the need for AI and next-generation AI applications grows, so does the need for computing power². According to a study by McKinsey, about 70% of the demand for gen AI compute will come from business-to-consumer (B2C) applications. This includes both simple interactions with customers (like writing emails) and more complex interactions with users (like turning text into images)².

What This Means for CPU Makers

The semiconductor business is getting close to a new S-curve². Leaders are responding by committing large amounts of money to expand data centers and semiconductor fabrication plants (fabs). At the same time, they are looking into new chip designs, materials, and architectures to meet the changing needs of a business world driven by generation AI².

The Challenge That Ahead

Costs go up with each new technology node because the money spent on research and design as well as capital equipment for production goes up a lot as structures get smaller². AI/ML is becoming an increasingly important tool across the whole value chain² as companies try to speed up time to market and boost productivity in research, chip design, and manufacturing.

The Future of Making CPUs

We expect there to be more inference-specific AI servers in 2030. These will use a mix of CPUs and several purpose-built AI accelerators that use ASICs². This move toward hardware that is designed to work with AI shows how AI is changing the industry that makes CPUs².

Bottom Line

The popularity of AI and the range of uses for it has had a big effect on CPU technology, forcing companies to change their designs to meet the needs of AI workloads⁴. As the AI revolution continues, CPU makers will have to come up with new ideas and change their products to keep up with the growing need for computing power⁹.

Learn more

1mckinsey.com

2mckinsey.com

3loginvsi.com

4saxon.ai

Leave a Comment

Your email address will not be published. Required fields are marked *