JP Data LLC
101 Jefferson Dr, 1st Floor
Menlo Park CA 94025

Phone
(408) 623 9165

Email
info at jpdata dot co
sales at jpdata dot co

AI chip companies continue to raise more capital

Note: This was published on Tractica web site in 2018.

High mask costs made it hard for chipset start-ups to raise capital during early 2000s. Today mask costs can range up to 25 million dollars and by some estimates the design costs for a chip at 12nm node can run as high as $174 million. This made it hard for investors to justify ROI as most of them demanded 10X return. Only a few markets offered high volume for chipsets to generate that kind of revenue.

With the emergence of AI, many new possibilities have opened up for AI chipsets by Tractica’s own estimation, the market is expected to reach $66 billion by 2025. The widespread applicability of AI application meant that many applications will need different chipsets leading to different requirements. This has led to many semiconductor and start-ups to jump in the market with their own solutions.

The start-ups started appearing in 2016 and as of 2018 and many cloud companies, top semiconductor companies, start ups, FPGA companies have announced their intention to make AI chips. Many of the seed rounds raised by start-ups were of the order of few milliions and that did nto get whole lot of attnetion. As of 2018, some of the companies have developed products and are able to command high valuation to chip companies. Cambricon, a China based ASIC company became the first official unicorn when it raised ‘hundreds of millions’ dollars recently to get total valuation of $2.5 billion. Cambricon is not the only company that is generating headlines. Many other companies have also raised a large capital. Some of the recent companies who have raised capitalinclude:

  • Wave Computing, a California company recently announced that they have raised $86 million in series E round.
  • Habana, who recently stormed on the ASIC scene by releasing its own chip for data center inferencing announced that it has raised $75 million.
  • Sambanova, a start-up out of UC Berkley raised $56 million
  • Groq, a start-up from the designers of TPU announced that it has raised $52 million in September 2018.
  • Thinci, a El Dorado, CA based company raised $65 million to build its products and solution in September 2018

The list goes on and on.  While some of these have released samples and products, some start-ups have been able to raise capital only on the basis of Powerpoint and team, something unheard of since late 90s when semiconductor companies were darling of the VC world.

Considering that the problem of AI hardware is hard, the investment is justified. A chip company not only has to manufacture a chip, but also provide customers with reference boards, IP, development tools and software to their customers. The software and IP carries more value than hardware for OEMs as very few people at this time have to skills to design AI applications from scratch. That means that semiconductor companies have to go extra mile to give a software and framework

AI chip companies continue to raise more capital

High mask costs made it hard for chipset start-ups to raise capital during early 2000s. Today mask costs can range up to 25 million dollars and by some estimates the design costs for a chip at 12nm node can run as high as $174 million. This made it hard for investors to justify ROI as most of them demanded 10X return. Only a few markets offered high volume for chipsets to generate that kind of revenue.

With the emergence of AI, many new possibilities have opened up for AI chipsets by Tractica’s own estimation, the market is expected to reach $66 billion by 2025. The widespread applicability of AI application meant that many applications will need different chipsets leading to different requirements. This has led to many semiconductor and start-ups to jump in the market with their own solutions.

The start-ups started appearing in 2016 and as of 2018 and many cloud companies, top semiconductor companies, start ups, FPGA companies have announced their intention to make AI chips. Many of the seed rounds raised by start-ups were of the order of few milliions and that did nto get whole lot of attnetion. As of 2018, some of the companies have developed products and are able to command high valuation to chip companies. Cambricon, a China based ASIC company became the first official unicorn when it raised ‘hundreds of millions’ dollars recently to get total valuation of $2.5 billion. Cambricon is not the only company that is generating headlines. Many other companies have also raised a large capital. Some of the recent companies who have raised capitalinclude:

  • Wave Computing, a California company recently announced that they have raised $86 million in series E round.
  • Habana, who recently stormed on the ASIC scene by releasing its own chip for data center inferencing announced that it has raised $75 million.
  • Sambanova, a start-up out of UC Berkley raised $56 million
  • Groq, a start-up from the designers of TPU announced that it has raised $52 million in September 2018.
  • Thinci, a El Dorado, CA based company raised $65 million to build its products and solution in September 2018

The list goes on and on.  While some of these have released samples and products, some start-ups have been able to raise capital only on the basis of Powerpoint and team, something unheard of since late 90s when semiconductor companies were darling of the VC world.

Considering that the problem of AI hardware is hard, the investment is justified. A chip company not only has to manufacture a chip, but also provide customers with reference boards, IP, development tools and software to their customers. The software and IP carries more value than hardware for OEMs as very few people at this time have to skills to design AI applications from scratch. That means that semiconductor companies have to go extra mile to give a software and framework

Nvidia today is de-factor leader in AI world, followed by Intel. Both companies are generating over billion dollars in revenue from AI chipset products. The value add of AI startups is in their hardware that offers power, performance benefits for a given application. Today a generic solution such as CPU and GPU is being used by most of the companies worldwide but the need for specialized chipset is widely recognized. Start ups are not quite shipping their products yet but when start ups start delivering the performance level they are promising, it will be an interesting battle.