Nvidia is constructing a new business division dedicated to creating custom chips, such as advanced artificial intelligence processors, for cloud computing companies and other enterprises.

The leading worldwide manufacturer and supplier of AI chips wants to secure a piece of the rapidly expanding custom AI chip market and defend itself against the increasing number of businesses looking to find alternatives to its goods.

The company, which is situated in Santa Clara, California, presently has roughly 80 percent of the market for high-end artificial intelligence processors. Due to this dominance, its market valuation has increased by 40 percent this year to $1.73 trillion, following a more than threefold increase in 2023.

In an attempt to compete in the quickly developing generative AI market, its clients, who include Microsoft, Alphabet, Meta Platforms, and OpenAI, the company that created ChatGPT, have rushed to seize the last of the Nvidia chips.

The chipmaker, which saw exponential growth in 2023, is concentrating not only on data center chips but also on industries like gaming, telecom, and automobiles.

According to research conducted by 650 Group and analyst Alan Weckel, the market for advance chips for data centers may grow to $10 billion by this year and perhaps quadruple by 2025.

On the other hand, Needham analyst Charles Shi projects that the global custom chip business would be worth approximately $30 billion by 2023, or roughly five percent of the total yearly semiconductor sales worldwide.

For many of those large customers, Nvidia's H100 and A100 chips function as a general-purpose, all-purpose AI processor. However, for certain purposes, IT companies have begun to design their own internal processors. By doing this, you can potentially cut down on energy consumption as well as design time and expense.

Nvidia said in 2022 that it would allow its customers to incorporate a portion of its exclusive networking technologies into their own chips.