(CTN News) – As per sources familiar with the company’s plans,is developing business unit dedicated to developing bespoke chips for cloud computing firms and others, including advanced artificial intelligence processors.
The dominant global designer and supplier of artificial intelligence (AI) chips is seeking to capture a portion of the exploding market for custom AI chips, and to protect itself from competing products.
The Santa Clara, California-based company currently holds about 80% of the market for high-end AI chips, a position which has sent its market value up 40% so far this year to $1.73 trillion after it tripled in 2023.
A number of its clients, including ChatGPT creator OpenAI, Microsoft, Alphabet, and Meta Platforms, have been racing to acquire Nvidia chips in order to compete in the rapidly emerging generative AI sector.
For many of these customers, Nvidia’s H100 and A100 chips serve as a general-purpose, all-purpose AI processor. It is true, however, that many technology companies have begun to develop their own internal chips for specific purposes.
Energy consumption can be reduced as a result of this method, as well as the cost and time associated with designing can be reduced.
In accordance with sources who declined to be identified due to their lack of authorization, Nvidia is now helping rival firms such as Broadcom and Marvell Technology develop custom artificial intelligence chips.
A system can’t afford to have an H100 or A100 thrown into it just because it has a lot of power or it it’s really expensive,” said Greg Reichow,Eclipse Ventures is a venture capital firm where I work as a general partner.
A good rule of thumb is to mix the right amount of compute with the right kind of computation if you want to achieve success.