Three AI companies created by post-90s generation, valued at 36 billion

AI financing is sweeping the world.

Recently, according to Reuters, Cohere, a star AI unicorn, has just raised $450 million from investors such as Nvidia, Salesforce Ventures, and Cisco, with an amazing valuation of $5 billion (RMB 36 billion).

You may not believe that the core founder is another post-95s - Aidan Gomez, who is also one of the authors of the famous AI benchmark paper "Attention is All You Need". At this year's Nvidia GTC conference, he was invited by Huang Renxun as a guest to participate in the roundtable dialogue.

From left to right: Ivan Zhang, Aidan Gomez, Nick Frosst

Since 2019, Aidan Gomez and two post-90s alumni, Ivan Zhang and Nick Frosst, founded Cohere and joined the AI ​​entrepreneurial wave. A newborn calf is not afraid of a tiger, which is undoubtedly the most vivid portrayal of AI making wealth. From this, we can see that a group of post-90s are leading the largest AI entrepreneurial wave in history.

Huang Renxun's guests

Three post-90s, creating a super unicorn

The story of Cohere began with three young people.

In 1995, Aidan Gomez was born in a forest village in Ontario, Canada. In his memory, his hometown is surrounded by green trees, beautiful mountains and clear waters, but the only thing is that the Internet speed is not good, and the Internet world seems out of reach, which drove him to learn about computers and fall in love with programming.

In 2013, Aidan entered the University of Toronto to study computer science, and then went to Oxford University for a doctorate. During this period, he extensively studied the academic papers of Geoffrey Hinton, a leader in the field of AI and a Turing Award winner, and even "criticized" the papers.

A turning point in his life was in 2017, when Aidan, who was in his early 20s, was interning at Google Brain and wrote and published the famous paper "Attention is All You Need" with seven other researchers. The paper first proposed the Transformer architecture, which was later the underlying architecture of the ChatGBT large model.

However, the birth of Transformer did not stir up too many waves at the time. Aidan watched the Transformer architecture being widely used within Google, driving incredible changes. But outside of Google, few people used Transformer to build applications. So he had the idea of ​​starting a business.

Coincidentally, at that time, Aidan met Ivan Zhang, a Chinese who dropped out of the University of Toronto to start a business. Ivan is an out-and-out technical practitioner, and the two hit it off-in 2019, Cohere, which focuses on the application of large AI models, came into being.

In 2020, another co-founder and chief scientist Nick Frosst joined. Nick also graduated from the Department of Computer Science at the University of Toronto and was a member of the Hinton team at Google Brain. So far, the three post-90s founders have assembled.

With his understanding of the underlying technology of large models, Aidan adopted a differentiated competitive strategy in his entrepreneurship. He did not take the involutionary "ChatGPT-like" route, but focused on using the capabilities of large models to serve corporate customers and designed a targeted payment model.

It is precisely because of this differentiated strategy that Cohere has gained a place in the fierce AI competition. According to foreign media reports, Cohere has currently cooperated with companies such as Jasper, HyperWrite, and Salesforce. As of the end of March this year, the company's revenue reached 35 million US dollars, with a gratifying growth.

At the NVIDIA GTC conference this year, the eight authors of the early papers reunited, and Aidan Gomez was also invited to be Huang Renxun's guest of honor. "Time" magazine released the first list of the world's top 100 AI figures, and Aidan Gomez was also among them.

One round of financing per year, with a valuation of 36 billion

Cohere's financing history has left a deep impression on the venture capital circle.

Thanks to Aidan's academic background, the company received support from bigwigs in its early days, including Turing Award winner Geoffrey Hinton, GAN father Ian Goodfellow, Uber Chief Scientist Raquel Urtasun, Nvidia Toronto Research Laboratory Director Sanja Fidler and Stanford Professor Fei-Fei Li. You know, these people are all leading figures in the current AI world.

With the endorsement of industry bigwigs, Cohere's financing rhythm is quite smooth, maintaining a rhythm of one round per year. In September 2021, Cohere completed a $40 million Series A financing from Index Ventures, Section 32, Radical Ventures, and several individual investors.

Six months later, in February 2022, Cohere received another $125 million in Series B financing, led by Tiger Global, with Radical Ventures, Index Ventures, and Section 32 following.

Until May 2023, Cohere announced the completion of a $270 million Series C financing, with a more luxurious lineup - led by Inovia Capital, Nvidia, Oracle, Salesforce, the global SaaS pioneer, Index Ventures, DTCP, Mirae Asset, Schroders Capital, SentinelOne, Thomvest Ventures, and South Korea's Future Asset Group. This brings the company's total financing to approximately $445 million, second only to OpenAI and Anthropic.

Since then, Cohere's valuation has exceeded 2 billion US dollars.

The latest scene is that Cohere has raised another 450 million US dollars from investors such as Nvidia, Salesforce, and Cisco, and its valuation has soared to 5 billion US dollars, which is astonishingly large.

A hot scene this year: Post-90s support the AI ​​entrepreneurship boom

Such a scene is not unfamiliar. Looking around, the post-90s are setting off an unprecedented AI entrepreneurship boom.

Behind every large amount of AI financing this year, you can almost see a group of post-90s academic founders. I still remember that last month, Scale AI announced that it had received $1 billion in financing, with a valuation of $13.8 billion (about RMB 100 billion), and the people standing behind them were Alexandr Wang and Lucy Guo, who were born after 1995. For details, please see Silicon Rabbit's previous article: A 1997 MIT dropout has become a $7.3 billion unicorn in five years.

Coincidentally, this week, AI automatic video generation software Pika completed a total of $80 million in Series B financing, led by Spark Capital, with Greycroft, Lightspeed Venture Partners and Jared Leto participating in the investment. The company's valuation exceeded $470 million (RMB 3.406 billion), and behind it was also a post-95s-Guo Wenjing.

Guo Wenjing was the first girl from Zhejiang to be admitted to Harvard undergraduate school in advance. She obtained a master's degree in computer science and an undergraduate degree in mathematics from Harvard and later went to Stanford University for a doctorate. In April last year, Guo Wenjing and several classmates dropped out of Stanford and jointly developed Pika, which became a global hit.

Looking at the domestic AI entrepreneurship boom, the younger generation is also coming in full force.

Not long ago, the latest round of financing for the domestic AI big model star company Dark Side of the Moon was leaked, and the valuation has reached 3 billion US dollars. The company leader is Yang Zhilin, a post-90s scholar from Tsinghua University. Growing up in Shantou, Guangdong, Yang Zhilin was recommended to Tsinghua University after graduating from high school, and later studied for a doctorate at Carnegie Mellon University. In the past year, Dark Side of the Moon has become a project that VCs are fiercely competing for in the big model track.

Also impressive is the former Huawei "genius boy" Zhihuijun Peng Zhihui. Graduated from the University of Electronic Science and Technology of China with a bachelor's and master's degree, he is also a B station up master with 2.5 million fans. The Zhiyuan Robot he founded is aimed at the field of embodied intelligence. Since its establishment in February 2023, Zhiyuan Robot has been working non-stop to win multiple rounds of financing and has been invincible.

There are more AI founders born in the 1990s. Zhang Yueguang, the former product manager of the hit AI product "Miaoya", founded Muyan Zhiyu, which has completed multiple rounds of financing in half a year. This post-90s graduate graduated from Tsinghua University; Jiang Yuchen, CEO of Waveform Intelligence, was born in 1998. He received his undergraduate degree from Zhejiang University and his doctorate from ETH Zurich. He is engaged in natural language processing research. Hangzhou Waveform Intelligence, founded by her, has recently received Pre-A round of financing led by LanChi Ventures, followed by Westlake Science and Technology Ventures and old shareholder Oufang Angel.

Here is another long list - Zeng Guoyang of Face Wall Intelligence, Hu Yuanming of Tai Chi Graphics, Tian Tian of Ruilai Wisdom, Chen Jianyu of Xingdong Jiyuan... Most of them have science and engineering degrees from top universities in China and even the world, and have become the darlings of the times with the help of artificial intelligence. In this regard, Huachuang Capital has summarized the characteristics of this wave of founders:

First, their academic backgrounds are quite outstanding, and they basically have strong related majors such as computer science and artificial intelligence in top domestic and foreign universities, and have rich research resources and internship experience in large companies;

Second, they focus on R&D rather than team, and only need a few people to build a complete team for development and implementation, and the start is very fast;

Third, generative AI is regarded as a huge paradigm shift, and young people have a strong belief and worship in technological innovation and Steve Jobs-style entrepreneurship.

This world will always belong to young people. Countless of them who graduated from prestigious universities are using the latest technology to write footnotes for the world and jointly create the magnificent scene of AI entrepreneurship in front of them.