Google CEO Sundar Pichai talks about the company’s third-generation artificial intelligence chips.
Source: YouTube screenshot
Some of the world’s largest technology companies are developing their own semiconductors, not content to rely on standard chips that are in high demand.
Apple, Amazon, Facebook, Tesla and Baidu shy away from established chip companies and bring certain aspects of chip development home, according to company announcements and media reports.
“Every time these companies want custom-made chips that fit the specific requirements of their applications instead of using the same generic chips as their competitors,” Syed Alam, world leader in Accenture semiconductors, told CNBC .
“This gives them more control over software and hardware integration while differentiating them from their competition,” Alam added.
Russ Shaw, former non-executive director of UK-based Semiconductor Dialog, told CNBC that custom-designed chips can perform better and run more economically.
“These specifically designed chips can help reduce the power consumption of specific technology company devices and products, whether it’s smartphones or cloud services,” Shaw said.
The current global shortage of chips is another reason why big tech companies are thinking twice about where they get their chips from, CNN’s Glenn O’Donnell, the firm’s research director, told CNBC. Forrester analysts. “The pandemic threw a big key in these supply chains, which accelerated efforts to get their own chips.”
“Many already felt limited in their pace of innovation by being locked into the timelines of chip makers,” O’Donnell said.
Artificial intelligence chips and more
Currently, just a month goes by without a Big Tech company announcing a new chip project.
Perhaps the most notable example was in November 2020 when Apple announced that it was moving away from Intel’s x86 architecture to make its own M1 processor, which is now on its new iMacs and iPads.
More recently, Tesla announced that it was building a “Dojo” chip to form artificial intelligence networks in data centers. The automaker in 2019 began producing cars with its custom AI chips that help integrated software make decisions in response to what’s happening on the road.
Baidu last month launched an AI chip designed to help devices process large amounts of data and increase computing power. Baidu said the “Kunlun 2” chip can be used in areas such as autonomous driving and that it has gone into mass production.
Some of the tech giants have chosen to keep certain semiconductor projects wrapped up.
According to reports, Google is nearing deploying its own CPUs or CPUs for its Chromebooks. According to a September 1 Nikkei Asia report, the search giant plans to use its CPUs on Chromebooks and tablets running the company’s Chrome operating system by 2023. Google did not immediately respond to a request for CNBC comments.
Amazon, which operates the world’s largest cloud service, is developing its own network chip to power hardware switches that move data over networks. If it works, it would reduce Amazon’s confidence in Broadcom. Amazon, which already designs other chips, did not immediately respond to a request for comment from CNBC.
Facebook’s chief AI scientist told Bloomberg in 2019 that the company is working on a new class of semiconductors that would work “very differently” than most existing designs. Facebook did not immediately respond to a request for comment from CNBC.
Design but not manufacture
At this stage, none of the tech giants are looking to do all the chip development on their own.
“It’s about the design and performance of the chip,” Shaw said. “At this stage, it’s not about manufacturing and casting, which costs a lot.”
Establishing an advanced chip factory, or foundry, such as the TSMC in Taiwan, costs about $ 10 billion and takes a few years.
“Even Google and Apple are reluctant to build them,” O’Donnell said. “They will go to TSMC or even Intel to build their chips.”
O’Donnell said there is a shortage of people in Silicon Valley with the skills needed to design high-end processors. “Silicon Valley put so much emphasis on software over the last few decades that hardware engineering was seen as a bit anachronistic,” he said.
“It got‘ cold ’to make hardware,” O’Donnell said. “Despite its name, Silicon Valley now employs relatively few real silicon engineers.”