Do you need a SPAC therapist?

Hello and welcome back to Equity, TechCrunch’s venture capital-focused podcast, where we unpack the numbers behind the headlines.

Natasha and Danny and Alex and Grace were all here to chat through the week’s biggest tech happenings. It was yet another busy week, but that just means we had a great time putting the show together and recording it. Honestly, we had a lot of fun this week, and we hope you crack a smile while we dig through the latest as a team.

Ready? Here’s the rundown:

  • The Coinbase direct listing! Here are our notes on its S-1, its direct listing reference price and its results. And we even wrote about the impact that it might have on other startup verticals!
  • Grab’s impending SPAC! As it turns out, Natasha loves SPACs now, and even Danny and Alex had very little to say that was rude about this one.
  • Degreed became a unicorn, proving yet again that education for the enterprise is a booming sub-sector.
  • Outschool also became an edtech unicorn, thanks to a new round led by Coatue and everyone’s rich cousin, Tiger Global. The conversation soon devolved into how Tiger Global is impacting the broader VC ecosystem, thanks to a fantastic analysis piece that you have to read here. 
  • Papa raised $60 million, also from Tiger Global. What do you call tech aimed at old folks? Don’t call it elder tech, we have a brand new phrase in store. Let’s see if it catches on.
  • AI chips! Danny talks the team through grokking Groq, so that we can talk about TPUs without losing our minds. He’s a good egg.
  • And, finally, Slice raised more money. Not from Tiger Global. We have good things to say about it.

And that is our show! We are back on Monday morning!

Equity drops every Monday at 7:00 a.m. PST, Wednesday, and Friday at 6:00 AM PST, so subscribe to us on Apple PodcastsOvercastSpotify and all the casts!

Google Cloud hires Intel veteran to head its custom chip efforts

There has been a growing industry trend in recent years for large scale companies to build their own chips. As part of that, Google announced today that it has hired long-time Intel executive Uri Frank as Vice President to run its custom chip division.

“The future of cloud infrastructure is bright, and it’s changing fast. As we continue to work to meet computing demands from around the world, today we are thrilled to welcome Uri Frank as our VP of Engineering for server chip design,” Amin Vahdat, Google Fellow and VP of systems infrastructure wrote in a blog post announcing the hire.

With Frank, Google gets an experienced chip industry executive, who spent more than two decades at Intel rising from engineering roles to Corporate Vice President at the Design Engineering Group, his final role before leaving the company earlier this month.

Frank will lead the custom chip division in Israel as part of Google. As he said in his announcement on LinkedIn, this was a big step to join a company with a long history of building custom silicon.

“Google has designed and built some of the world’s largest and most efficient computing systems. For a long time, custom chips have been an important part of this strategy. I look forward to growing a team here in Israel while accelerating Google Cloud’s innovations in compute infrastructure,” Frank wrote.

Google’s history of building its own chips dates back to 2015 when it launched the first TensorFlow chips. It moved into video processing chips in 2018 and added OpenTitan , an open source chip with a security angle in 2019.

Frank’s job will be to continue to build on this previous experience to work with customers and partners to build new custom chip architectures. The company wants to move away from buying motherboard components from different vendors to building its own “system on a chip” or SoC, which it says will be drastically more efficient.

“Instead of integrating components on a motherboard where they are separated by inches of wires, we are turning to “Systems on Chip” (SoC) designs where multiple functions sit on the same chip, or on multiple chips inside one package. In other words, the SoC is the new motherboard,” Vahdat wrote.

While Google was early to the ‘Build Your Own Chip’ movement, we’ve seen other large scale companies like Amazon, Facebook, Apple and Microsoft begin building their own custom chips in recent years to meet each company’s unique needs, and give more precise control over the relationship between the hardware and software.

It will be Frank’s job to lead Google’s custom chip unit and help bring it to the next level.

Qualcomm-backed chipmaker Kneron nails Foxconn funding, deal

A startup based out of San Diego and Taipei is quietly nailing fundings and deals from some of the biggest names in electronics. Kneron, which specializes in energy-efficient processors for edge artificial intelligence, just raised a strategic funding round from Taiwan’s manufacturing giant Foxconn and integrated circuit producer Winbond.

The deal came a year after Kneron closed a $40 million round led by Hong Kong tycoon Li Ka-Shing’s Horizons Ventures. Amongst its other prominent investors are Alibaba Entrepreneurship Fund, Sequoia Capital, Qualcomm and SparkLabs Taipei.

Kneron declined to disclose the dollar amount of the investment from Foxconn and Winbond due to investor requests but said it was an “eight figures” deal, founder and CEO Albert Liu told TechCrunch in an interview.

Founded in 2015, Kneron’s latest product is a neural processing unit that can enable sophisticated AI applications without relying on the cloud. The startup is directly taking on the chips of Intel and Google, which it claims are more energy-consuming than its offering. The startup recently got a talent boost after hiring Davis Chen, Qualcomm’s former Taipei head of engineering.

Among Kneron’s customers are Chinese air conditioning giant Gree and German’s autonomous driving software provider Teraki, and the new deal is turning the world’s largest electronics manufacturer into a client. As part of the strategic agreement, Kneron will work with Foxconn on the latter’s smart manufacturing and newly introduced open platform for electric vehicles, while its work with Winbond will focus on microcontroller unit (MCU)-based AI and memory computing.

“Low-power AI chips are pretty easy to put into sensors. We all know that in some operation lines, sensors are quite small, so it’s not easy to use a big GPU [graphics processing unit] or CPU [central processing unit], especially when power consumption is a big concern,” said Liu, who held R&D positions at Qualcomm and Samsung before founding Kneron.

Unlike some of its competitors, Kneron designs chips for a wide range of use cases, from manufacturing, smart home, smartphones, robotics, surveillance, payments, to autonomous driving. It doesn’t just make chips but also the AI software embedded in the chips, a strategy that Liu said differentiates his company from China’s AI darlings like SenseTime and Megvii, which enable AI service through the cloud.

Kneron has also been on a less aggressive funding pace than these companies, which fuel their rapid expansion through outsize financing rounds. Six-year-old SenseTime has raised about $2.6 billion to date, while nine-year-old Megvii has banked about $1.4 billion. Kneron, in comparison, has raised just over $70 million from a Series A round.

Like the Chinese AI upstarts, Kneron is weighing an initial public offering. The company is expected to make a profit in 2023, Liu said, and “that will probably be a good time for us to go IPO.”

Horizon Robotics, a Chinese rival to Nvidia, seeks to raise over $700M

In their rush to offer alternatives to advanced western chipsets, Chinese semiconductor companies are racking up large fundings from investors. Horizon Robotics, a five-year-old unicorn specializing in AI chips for robots and autonomous vehicles, announced Tuesday that it has secured $150 million in funding.

The proceeds are the first close of an over $700 million Series C round that Horizon is seeking to raise. The partial funding is jointly led by prominent investors 5Y Capital (formerly Morningside Venture Capital), Hillhouse Capital, and Capital Today. Chinese brokerage Guotai Junan’s international arm and KTB.

The round arrived less than two years after Horizon completed its $600 million Series B round, which valued the firm at $3 billion post-money and saw the participation of prominent Korean financiers including SK China, the China subsidiary of conglomerate SK Group, and SK Hynix, SK’s semiconductor unit.

The startup, founded by a Baidu veteran, raised its Series A round of over $100 million led by Intel Capital in late 2017.

 

With the fresh capital, Horizon plans to hasten the development and commercialization of its automotive chips and autonomous driving solutions. It also aims to build an “open ecosystem” for industry partners.

For the past couple of years, China has been striving to wean dependence on western chip giants in sectors ranging from smartphones to vehicles. Local startups like Horizon Robotics and Black Sesame Technologies, as well as telecoms titan Huawei, are pouring resources into autonomous driving processors, hoping to match or overtake the technologies from Nvidia and Intel’s Mobileye.

Horizon’s OEM and Tier 1 auto partners, according to the firm, include Audi, Bosch, Continental, SAIC Motor and BYD.

75% of China’s ADAS (advanced driver-assistance system)-equipped cars and Level 3 (autonomous driving under certain circumstances) vehicles will be supported by Chinese suppliers by 2030, up from 20% in 2019, investment bank CITIC Securities projects.

Kneron launches its new AI chip to challenge Google and others

Fresh off a $40 million Series A round, edge AI specialist Kneron today announced the launch of its newest custom chip, the Kneron KL 720 SoC.

With funding from the likes of Alibaba, Sequoia, Horizons Ventures, Qualcomm and SparkLabs Taipei (as well as a few undisclosed backers), it’s worth taking the company’s efforts seriously, and Kneron has no qualms about comparing its chips to those of Intel and Google, for example. It argues that its KL 720 is twice as energy efficient as Intel’s latest Movidius chips and four times more efficient than Google’s Coral Edge TPU at running the MobileNetV2 image recognition benchmark.

Compared to its previous generation of chips, this updated version can process 4K still images and videos at a 1080P resolution. It also features a number of new audio recognition breakthroughs for the company, which Kneron says will allow devices that use its chips to bypass the standard wake words on other chips and have immediate conversations with the device.

Image Credits: Kneron

Overall, Kneron promises 1.5 TOPS in performance from its SoC, which uses an Arm Cortex M4 as its main control unit. The average power consumption for the full package is around 1.2W.

“KL720 combines power with unmatched energy-efficiency and Kneron’s industry-leading AI algorithms to enable a new era for smart devices,” said Kneron founder and CEO Albert Liu. “Its low cost enables even more devices to take advantage of the benefits of edge AI, protecting user privacy, to an extent competitors can’t match. Combined with our existing KL520, we are proud to offer the most comprehensive suite of AI chips and software for devices on the market.”

With KNEO, the company also offers an interesting networking solution for devices that are powered by its chips. With this, developers can create their own private networks and connect multiple sensors without having to route data to the cloud. That network uses blockchain technology to secure the data and in a bit of a twist, Kneron hopes to create a marketplace that will allow consumers to exchange or sell their data to buyers.

For now, though, the company seems to be more focused on the core hardware. That’s also an area where we’ve seen the competition heat up, with other well-funded startups like Hailo also recently launching their latest chips.

Wendell Brooks has resigned as president of Intel Capital

When Wendell Brooks was promoted to president of Intel Capital, the investment arm of the chip giant, in 2015, he knew he had big shoes to fill. He was taking over from Arvind Sodhani, who had run the investment component for 28 years since its inception. Today, the company confirmed reports that Brooks has resigned that role.

Wendell Brooks has resigned from Intel to pursue other opportunities. We thank Wendell for all his contributions and wish him the best for the future,” a company spokesperson told TechCrunch in a rather bland send off.

Anthony Lin, who has been leading mergers and acquisitions and international investing, will take over on an interim basis. Interestingly, when Brooks was promoted, he too was in charge of mergers and acquisitions. Whether Lin keeps that role remains to be seen.

When I spoke to Brooks in 2015 as he was about to take over from Sodhani, he certainly sounded ready for the task at hand. “I have huge shoes to fill in maintaining that track record,” he said at the time. “I view it as a huge opportunity to grow the focus of organization where we can provide strategic value to portfolio companies.”

In that same interview, Brooks described his investment philosophy, saying he preferred to lead, rather than come on as a secondary investor. “I tend to think the lead investor is able to influence the business thesis, the route to market, the direction, the technology of a startup more than a passive investor,” he said. He added that it also tends to get board seats that can provide additional influence.

Comparing his firm to traditional VC firms, he said they were as good or better in terms of the investing record, and as a strategic investor brought some other advantages as well. “Some of the traditional VCs are focused on a company-building value. We can provide strategic guidance and complement some of the company building over other VCs,” he said.

Over the life of the firm, it has invested $12.9 billion in more than 1,500 companies, with 692 of those exiting via IPO or acquisition. Just this year, under Brooks’ leadership, the company has invested $225 million so far, including 11 new investments and 26 investments in companies already in the portfolio.

Cambricon, once Huawei’s core AI chip supplier, eyes $400M IPO

One of China’s most valuable artificial intelligence chipmakers Cambricon is one step closer to its initial public offering, and its prospectus reveals a rare snapshot of where Chinese companies stand in relation to their international counterparts in this critical field.

Cambricon got the nod in early June to list on the Star Market, China’s new Nasdaq-like stock exchange conceived to attract high-potential tech startups. This week, the chipmaker received the final green light from the China Securities Regulatory Commission, the stock market watchdog, for its first-time sale.

The company is aiming to raise 2.8 billion yuan ($400 million) from its IPO and spend the proceeds on cloud-based algorithm training and inference, edge computing, and cash flow boost. It was last valued at 2.5 billion yuan in 2018 and expects its market cap to exceed 1.5 billion yuan when it floats.

Cambricon began life in a lab within the Chinese Academy of Sciences (CAS), the national institute for science and technology backed by government money. In 2016, the project spun out as a separate entity, making money by licensing intellectual property and selling chips for deep-learning acceleration. Before long, it had made its name as a major supplier of Huawei’s first AI chip-powered smartphones and other flagship models later on.

But the partners’ ties have weakened ever since Huawei began doubling down on its own semiconductor arm — HiSilicon — to hedge against U.S. sanctions. The direct consequence is a substantial revenue drop for Cambricon’s licensable IP, which slumped to an estimated 16-18 million yuan in 2018, down from 117 million yuan in 2018.

“Huawei Silicon has chosen to develop its own AI chips for end devices and has not extended the partnership with our company, and our AI chip business with other clients remains relatively small,” the company replied to regulators during the vetting process for its listing. Finding new clients at Huawei’s enormous scale is also challenging, as “most of the other well-known Chinese smartphone makers are using established handset chips and solutions from Qualcomm and MediaTek,” Cambricon noted.

The chipmaker also flagged that it remains “well behind” international competitors such as Nvidia, Intel, AMD in areas including “overall scale, capital reserve, resources for research and development and sales channels.” It’s also well aware of rising domestic competition from its old ally, Huawei, which has opted for chips from its home-grown HiSilicon unit.

Cambricon’s co-founders Chen Tianshi and Chen Yunji both hail from academia. The company still maintains close relationships with CAS and also works closely with Olivier Temam, a researcher at Inria, the French national institute for computer science and applied mathematics.

Cambricon is still operating in the red, adding up to a total loss of 1.6 billion yuan ($230 million) in the last three years in part due to large sums spent on research and development, according to its prospectus. It generated revenues of 444 million yuan ($63 million) in 2019, up from 7.84 million yuan in 2017.

The chipmaker is backed by a lineup of storied investors across the board. Besides the 41.7% stake Chen Tianshi commands, other shareholders include Zhongke Suanyuan, an asset management firm set up by CAS; Aixi Partners, an entity owned by Cambricon employees and controlled by Chen Tianshi; SDIC Venture Capital, a state-owned investment firm approved by China’s state council; e-commerce titan Alibaba; and voice recognition provider iFlytek.

White House reportedly in talks with Intel, TSMC to build advanced chip foundries in the U.S.

White House officials are talking to Intel and TSMC about building semiconductor foundries in the United States, according to a Wall Street Journal report. U.S. tech companies and the government have been trying to reduce the country’s dependence on chip factories in Asia for years, underscored by national security concerns, the U.S.-China tariff war and now the COVID-19 pandemic, which has disrupted supply chains and logistics around the world.

The WSJ also reported that some U.S. officials have also talked to Samsung Electronics about expanding its existing contract-manufacturing operations in the U.S. to produce more advanced chips.

Intel, TSMC and Samsung Electronics are able to make chips of 10-nanometers or lower, the fastest and most power-efficient chips currently on the market.

In an April 28 letter obtained by the WSJ, Intel CEO Bob Swan told Defense Department that the company is willing to build a commercial foundry in partnership with the Pentagon “given the uncertainty created by the current geopolitical situation.”

Intel already has U.S. operations that make chips for its own products, but the new factory would serve other companies as well. TSMC, a Taiwanese semiconductor contract manufacturer, would continue making chips for other companies (its customers include Qualcomm, Nvidia and Advanced Micro Devices).

The newspaper reports that TSMC has been in talks with Commerce and Defense department officials and Apple, one of the biggest clients, about building a semiconductor factory in the U.S. The company said it is considering opening an overseas plant, but has finalized a decision yet.

“We are actively evaluating all suitable locations, including the U.S., but there is no concrete plan yet,” a TSMC spokesperson told the WSJ.

Other solutions that have been proposed by U.S. officials and industry groups include government investment in the domestic chip industry to support the high cost of building foundries, tax credits for semiconductor makers to buy and install equipment at U.S factories, and implementing more export restrictions for U.S. companies that ship microchips to buyers in China.

TechCrunch has contacted Intel and TSMC for comment.

Alibaba unveils Hanguang 800, an AI inference chip it says significantly increases the speed of machine learning tasks

Alibaba Group introduced its first AI inference chip today, a neural processing unit called Hanguang 800 that it says makes performing machine learning tasks dramatically faster and more energy-efficient. The chip, announced today during Alibaba Cloud’s annual Apsara Computing Conference in Hangzhou, is already being used to power features on Alibaba’s e-commerce sites, including product search and personalized recommendations. It will be made available to Alibaba Cloud customers later.

As an example of what the chip can do, Alibaba said it usually takes Taobao an hour to categorize the one billion product images that are uploaded to the e-commerce platform each day by merchants and prepare them for search and personalized recommendations. Using Hanguang 800, Taobao was able to complete the task in only five minutes.

Alibaba is already using Hanguang 800 in many of its business operations that need machine processing. In addition to product search and recommendations, this includes automatic translation on its e-commerce sites, advertising and intelligence customer services.

Though Alibaba hasn’t revealed when the chip will be available to its cloud customers, the chip may help Chinese companies reduce their dependence on U.S. technology as the trade war makes business partnerships between Chinese and American tech companies more difficult. It can also help Alibaba Cloud grow in markets outside of China. Within China, it is the market leader, but in the Asia-Pacific region, Alibaba Cloud still ranks behind Amazon, Microsoft and Google, according to the Synergy Research Group.

Hanguang 800 was created by T-Head, the unit that leads the development of chips for cloud and edge computing within Alibaba DAMO Academy, the global research and development initiative that Alibaba is investing more than $15 billion in. T-Head developed the chip’s hardware and algorithms designed for business apps, including Alibaba’s retail and logistics apps.

In a statement, Alibaba Group CTO and president of Alibaba Cloud Intelligence Jeff Zhang (pictured above) said “The launch of Hanguang 800 is an important step in our pursuit of next-generation technologies, boosting computing capabilities that will drive both our current and emerging businesses while improving energy-efficiency.”

He added “In the near future, we plan to empower our clients by providing access through our cloud business to the advanced computing that is made possible by the chip, anytime and anywhere.”

T-Head’s other launches included the XuanTie 910 earlier this year, an IoT processor based on RISC-V, the open-source hardware instruction set that began as a project at U.C. Berkeley. XuanTie 910 was created for heavy-duty IoT applications, including edge servers, networking, gateway and autonomous vehicles.

Alibaba DAMO Academy collaborates with universities around the world that have included U.C. Berkeley and Tel Aviv University. Researchers in the program focus on machine learning, network security, visual computing and natural language processing, with the goal of serving two billion customers and creating 100 million jobs by 2035.

Google and Twitter are using AMD’s new EPYC Rome processors in their datacenters

AMD announced that Google and Twitter are among the companies now using EPYC Rome processors during a launch event for the 7nm chips today. The release of EPYC Rome marks a major step in AMD’s processor war with Intel, which said last month that its own 7nm chips, Ice Lake, won’t be available until 2021 (though it is expected to release its 10nm node this year).

Intel is still the biggest datacenter processor maker by far, however, and also counts Google and Twitter among its customers. But AMD’s latest releases and its strategy of undercutting competitors with lower pricing have quickly transformed it into a formidable rival.

Google has used other AMD chips before, including in its “Millionth Server,” built in 2008, and says it is now the first company to use second-generation EPYC chips in its datacenters. Later this year, Google will also make virtual machines that run on the chips available to Google Cloud customers.

In a press statement, Bart Sano, Google vice president of engineering, said “AMD 2nd Gen Epyc processors will help us continue to do what we do best in our datacenters: innovate. Its scalable compute, memory and I/O performance will expand out ability to drive innovation forward in our infrastructure and will give Google Cloud customers the flexibility to choose the best VM for their workloads.”

Twitter plans to begin using EPYC Rome in its datacenter infrastructure later this year. Its senior director of engineering, Jennifer Fraser, said the chips will reduce the energy consumption of its datacenters. “Using the AMD EPYC 7702 processor, we can scale out our compute clusters with more cores in less space using less power, which translates to 25% lower [total cost of ownership] for Twitter.”

In a comparison test between 2-socket Intel Xeon 6242 and AMD EPYC 7702P processors, AMD claimed that its chips were able to reduce total cost of ownership by up to 50% across “numerous workloads.” AMD EPYC Rome’s flagship is the 64-core, 128-thread 7742 chip, with a 2.25 base frequency, 225 default TDP and 256MB of total cache, starts at $6,950.