Silicon Valley's old money and Chinese giants, AI M&A tide peers have different fates

Source: Light Cone Intelligence, Authors: Yao Yue, Hao Xin

Image source: Generated by Unbounded AI‌

ChatGPT kicked off a new round of AI venture capital upsurge, but in China and Silicon Valley, it presents a "two worlds of ice and fire".

In China, light-years away from the large-scale start-up company that has just received financing, because the founder Wang Huiwen was in ill health, he was urgently taken over for 1 yuan; The largest merger and acquisition in the field-data giant Databricks acquired AI startup MosaicML for US$1.3 billion.

What the hot and cold capital market reflects is that China's AIGC industry is in the early stages of development. Faced with unclear business models and insufficient technology, a large number of domestic entrepreneurs still have a long way to go.

Wang Xing made a move, and took over from the upper bunk brothers

Wang Huiwen left the post due to illness, and was quickly taken over by Meituan Light Years Away.

** On June 29, Meituan announced that it had signed an agreement to fully acquire Light Years Beyond. The consideration includes paying 230 million US dollars (approximately 1.668 billion yuan) for the purchase of overseas equity, 367 million yuan of domestic debt, and 1 yuan in cash. **

(Source: Meituan Announcement)

The total price was 2.035 billion yuan, which was immediately called out as the largest AIGC acquisition in China this year. But in fact, there is still 285 million US dollars (about 2.067 billion yuan) in cash on the account light years away. "This is equivalent to selling it for 1 yuan, with no premium. You can't see any value of the company from the transaction." Huang Lichong, president of Huisheng International Capital, said.

It is worth mentioning that, according to IT Juzi data, just on June 5, Light Years Away just received a strategic investment of 230 million US dollars (about 1.668 billion yuan), and the post-investment valuation reached 1.23 billion US dollars (about 1.668 billion yuan). 8.92 billion yuan). This amount plus the 50 million US dollars (approximately 363 million yuan) brought by Wang Huiwen is almost the same as the cash in the account. It can be seen that beyond the light years of being promoted to a unicorn, there is an accident before spending money to do more things to show value.

**Technology tycoon Wang Huiwen and the brothers and capital behind him who supported him failed to succeed, which is embarrassing. **

Just a few months ago, Wang Huiwen was in command "with capital" and was backed by a luxurious VC team. The outside world has always maintained high expectations for light years away, and now his fate is naturally attracting attention.

Based on the fact that it has been acquired by Meituan, there is an optimistic guess that Wang Xing was already interested in the big model before it became popular, and Meituan has also expanded its algorithm team. The direction of AGI (General Artificial Intelligence) insisted by Wang Huiwen is also very suitable for the business attributes of Meituan’s strong ToC. For light-years away, Meituan is worth entrusting.

At present, among the first echelon of large technology companies, only Meituan has not announced the related actions of the large model.

But if you take a closer look at the matter, you will find that this merger, which was ostensibly made by Meituan, is actually a "dissolution".

Those who are familiar with the primary market will see that the current US$280 million in cash from light years away comes from US$50 million in debt (most likely the convertible bonds given by Wang Huiwen), and the consideration for the acquisition of Meituan is US$280 million. It is equivalent to returning the investment funds and convertible bonds, and the company's enterprise value is worth 1 yuan. In essence, it is to dissolve the company through the acquisition of Meituan.

The current whereabouts of the most important core technical personnel is still unknown.

This also seems to be an important footnote for a recent debate that has attracted much attention in the AIGC circle.

AIGC started a business, met "Silicon Valley old money" and "Chinese giants"

A few days ago, Zhu Xiaohu, the managing partner of GSR Ventures, and Fu Sheng, the CEO of Cheetah Mobile, had a debate on whether the big model has entrepreneurial value in the circle of friends, which aroused great concern in the circle. Underneath the superficial fanaticism of the large model, the current situation full of contradictions was pushed to the surface.

(Source: Fu Sheng Moments)

As a representative company of domestic large-scale entrepreneurship, the situation beyond light years undoubtedly aggravates the market's doubts about domestic large-scale entrepreneurship and even AIGC's entrepreneurship.

From the data point of view, the recent primary market is indeed quite cautious. According to Qichacha, there were 456 public investment and financing incidents in the domestic artificial intelligence industry in the first half of 2023. And this statistic is 731, 526, 353, 631 and 648 in the five years from 2018 to 2022. Only the first half of 2020, when the epidemic was the worst, was lower than this year.

Yang Zhen, an industry investor, also told Guangcone Intelligence: "There are many people looking at the AI market, but few people investing. As an investor, the most important thing is to invest in reasonable AI projects at a reasonable time. I invest in AI projects myself. The standard is to give priority to projects in the pure incremental market, which can accumulate data and expand scenarios based on the incremental market.”

In contrast, Silicon Valley is a different story. Within one day on June 26 alone, three mergers and acquisitions were announced in the AIGC field.

Most notably, big data giant Databricks acquired AI startup MosaicML at a high price of US$1.3 billion, which became the highest acquisition amount in the AIGC field this year. Databricks has met the rapidly growing need for enterprises to build ChatGPT-like tools through acquisitions.

(Source: MosaicML official website)

Financial automation company Ramp will acquire OpenAI rival Cohere.io, which has implemented AIGC and LLM (big language model) to automate bill processing; business and intelligent information provider Thomson Reuters will acquire AI legal technology startup Casetext, Under the core product with AI to review document functionality.

Only three days later, on June 29, AI start-up company Inflection announced the completion of US$1.3 billion in financing. This round of financing was led by Microsoft, Nvidia, etc., and its total financing reached US$1.525 billion. After this round of financing, Inflection AI will become the second largest startup in the field of artificial intelligence after Open AI.

(Source: Inflection official website)

Not only that, traditional technology giants IBM and cloud computing giant Snowflake have made acquisitions, and foreign companies have come forward to express "openness to the acquisition of AI start-ups."

According to PitchBook data, since the beginning of this year, the median pre-money valuation of AIGC’s early-stage financing has soared by 16% compared with last year, while the A or B round financing prices of other start-up companies have dropped by nearly 24%.

**This is a big gamble that the "old money" in Silicon Valley is betting on. **

Microsoft invested in OpenAI and got the AGI ticket. While other technology companies generally face difficulties such as sluggish revenue growth and stagnant old businesses, AIGC may not be able to solve all problems, but the future outlined is quite attractive.

The direction of the giant wheel is far less flexible and rapid than that of a start-up company. Therefore, through acquisitions and investments, we can make up for the shortcomings of AI, upgrade old products, and hope to regain market confidence.

As Xinyi Capital Qian Kaikai said, "In the AI2.0 stage, the barriers built by many established software companies have loosened, but it is difficult to reinvent themselves, which gives opportunities to start-up companies."

However, these veterans are also very familiar with the essence of "bringing doctrine": not every company needs to establish its own ChatGPT, and it is enough to combine AI technology with industry knowledge and data.

Looking at the acquired and invested companies with the standards of old money, you will find that they all meet at least four conditions, namely, technology, business 2B, product scenarios, and commercialization potential.

On the surface, the acquired and invested start-up companies have achieved a "flying rise". For example, the valuation of MosaicML has jumped six times. Runway, which is regarded as the strongest competitor of Midjourney and Stable diffusion, has received a US$100 million investment from Google. rose to $1.5 billion.

But at a deeper level, this may be the foresight of the old money, replacing cash acquisitions with shares, or acquiring at a price far exceeding the market, so as to achieve the purpose of defending the company's valuation. AI tools” news, the stock prices of these companies can rise a wave.

In contrast, the protagonists of China's large-scale model industry are still leading technology giants such as Ali, Baidu, Tencent, and Byte, and vertical giants in the AI field. Most entrepreneurs in the middle waist still choose to focus on the application direction There is no subversive innovation in "successful success".

And these giants are more inclined to end by themselves, rather than mergers and acquisitions, investing in start-up companies. This scene is very similar to the mobile Internet ten years ago.

China's AIGC road is blocked and long

"Money is the smartest thing." The capital market is hot and cold, which is why there is a certain gap between China and the US AIGC.

The emergence of foreign large-scale model mergers and acquisitions has released a strong industry signal, which indicates that the development of foreign AIGC has entered a new stage, and technology, business, scenarios, and commercialization will continue to advance.

In the domestic primary market, investors are cautiously optimistic and there are multiple complicated factors behind it. Some investors said that at present, it is difficult for AIGC to impress investors in C-end applications because of strong personalization and too many uncontrollable factors; B-end applications face slow digitization at the enterprise end, weak willingness to pay for software, and slow commercialization . **

AIGC is more closely integrated with the underlying technology. Although it has been struggling to catch up, there is still a big gap in AI technology, which has a strong support for commercialization and is weak.

In a recent online interview with a local online media in the United States, Musk said that the actual gap between China and the United States in the field of AI is about 12 months.

In the eyes of the Silicon Valley venture capital fund Leonis Capital, the gap is even wider. "Chinese developers can always catch up with Western leaders, but it will take 2-3 years for this technology to catch up."

But at the same time, we must also see that China is the only country outside the US market that has a complete AIGC industrial chain. **And the AIGC industry in the United States has become a red ocean. For example, dozens of start-up companies are already working on text generation products in the field of copywriting and marketing; while most of the AIGC industry in China is still a blue ocean, and some fields have not even There are a lot of startups popping up.

Wang Huiwen left the large-scale model battlefield, and there are still a large wave of entrepreneurs still struggling to advance. "There will be surprises in the journey of China's large-scale model catching up with the top international level," recently, Jiao Keru, the co-founder of Baichuan Intelligent, said.

AIGC has a long and difficult road to landing, and entrepreneurs need to be mentally prepared for a protracted battle. I hope all entrepreneurs take care of their health, and I wish Wang Huiwen a speedy recovery.