Investors "roll" to AI

Source: Burning Dimensions (ID:chaintruth), Author: Ma Shuye, Editor: Xie Zhongxiu

Image source: Generated by Unbounded AI tool

The AIGC upsurge continues to "burn", and it also affects and changes domestic investors in all aspects.

"After arriving at the company, first reply to the email, and then check the industry information and financing information. In particular, new concepts appear on the AIGC track every day. In addition, you need to read the latest academic papers in international academic journals. The most workload is to check the industry Research report (referred to as research report).” Wang Xiaohui, managing partner of Shengjing Jiacheng Investment, described to Ranjiyuan his work rhythm every morning.

Previously, if he wanted to obtain data of a certain industry scale, Wang Xiaohui had to read at least 5-6 research reports with dozens of pages at a time. In order to provide data support for the project, investors often need to spend more time looking for available data in the huge research report library.

Now "training AI to work for itself", AIGC provides investors with efficiency tools.

"Now it's all AI+. I will use ChatGPT to focus on research reports and train ChatGPT to work for me." Wang Xiaohui said to Ranjiyuan that from the end of 2022 to the present, AI tools have been used to capture industry research reports and generate mind maps. Let her The team's work efficiency doubled. In addition, using ChatGPT to output the basic code and undertake the design requirements also allows Wang Xiaohui to "save worry and effort" and save costs.

Like Wang Xiaohui, Xiao, who is currently working in an overseas US dollar fund, also said that in overseas workplaces, AI tools are still used to correct grammatical problems and correct expressions. , Buffett and other roles to assist investment decision-making".

In addition to auxiliary work, AIGC fever has also brought about changes in the focus of investors.

At present, major domestic companies such as Baidu, Tencent Youtu, Alibaba, Kuaishou, ByteDance, NetEase, SenseTime, Meitu, etc. have entered the AIGC track, including Baidu’s Wenxinyiyan and Tencent’s writing robot "" "Dream Writer", Alibaba's AI online design platform Lubanner, etc. Later, AI leaders such as Wang Huiwen, Wang Xiaochuan, Li Zhifei, etc. also brought up AI large-scale model entrepreneurship fever, AIGC has become an investment that investors have to pay attention to hotspot areas.

Investors flocked in, and the AIGC investment track also had a situation of "good projects don't ask for investment". A senior investor told Ranjiyuan, "Some investor friends complained to me that Wang Xiaochuan made a large model, and he took a large sum of money to invest and was rejected."

**At the same time, investors are looking down for financing opportunities for good projects, but on the other hand, they also need to be wary of the AIGC entrepreneurs who are being swayed. , claiming that the technical level is far superior to those that have been put into use, but after investigation, it was found that they just copied ChatGPT’s open source software.”

"Many start-up companies now connect to ChatGPT in the backend, make a UI design on the front end, and put it on the Apple store to promote that they are doing AIGC entrepreneurship." William Wong, managing director of Unknown Capital, said that this kind of "pipe skin" AIGC project, "no Technical barriers and business logic are just hype."

Google CEO Eric Schmidt once said, "If someone offers you a spot on a rocket, don't care if the spot is good or bad, go up and talk about it first."

From web3 in 2022 to AIGC in 2023, investors are scrambling to get on the "express train" of AIGC. After all, under the upsurge, no one wants to be left behind.

But at present, there are still many challenges and problems to be solved on the domestic AIGC track. Even investors who are eager to step on the express train of AIGC are very cautious in "shooting". According to the data of Qichacha, as of the end of May, only 5 related financings in the field of AI text/image/video generation have been completed in 2023.

Under the turmoil, investors are also inclined to be rational. "The concept of AIGC in China is still in the early stage. We want to see whether there are still entrepreneurs in the market after the turmoil recedes." William added.

Squeeze into the AI track, muddy and sandy

Under the upsurge, investors have set their sights on the AI track.

"Investors and entrepreneurs all over Silicon Valley are developing in this direction." Jiang Tao, founder of CSDN and founding partner of Geekbang Venture Capital, said bluntly at the Pioneer Conference of Artificial Intelligence Conference (GAIDC).

Under the upsurge, the AIGC investment track is also hot, good projects have to be "grabbed", and there are many projects that "get hot" under the mud and sand.

"There is a lot of hype, and there are too few good projects. When encountering good projects, investors often rely on 'grabbing'." Wang Xiaohui told Ranjiyuan.

As early as 2018, Wang Xiaohui was already paying attention to the AIGC field. Even if she made an early deployment, she could also feel that as ChatGPT triggered heated discussions at the end of 2022, the investment projects driving the AI track were completely "robbed". A good project that can finally come out has become the goal of investors.

Wang Xiaohui added, "In any industry, there are not many good projects, and good projects do not have to worry about financing."

Even, as early as 2018, when Wang Xiaohui participated in an exchange activity of a scientific research institute, he paid attention to the research sharing about "AI for science" shared by Zhang Linfeng, who was still a doctoral student at that time, "His research was at that time, and even now They are all very peak technological innovations, so I expressed my investment intention very early, but he had not returned to China at that time.”

Wang Xiaohui said that after Zhang Linfeng returned to China to establish Shenshi Technology during the epidemic, good projects also triggered competition for other large capitals. Finally, after several rounds of competition, Wang Xiaohui invested in Shenshi Technology in the angel round. After that, Shenshi Technology won It has received multiple rounds of investment from institutions such as Hillhouse, Matrix Partners, Qiming, and Huawei Hubble.

In this regard, Zhang Jinjian, founding partner of Oasis Capital, said that as a large number of start-up companies flooded into the AIGC track, as of March 2023, in the past three weeks, the Oasis team can only sleep for four or five hours a day, depending on how close they are. 20 items.

Ranjiyuan searched for platforms such as "Activities", ChatGPT and AIGC topped the hot search list, and "AIGC summits" large and small also had the presence of major investment institutions.

Figure/Activity row search page source/Screenshot of Burning Dimensions

Xiao said that in the financial field, there will always be new hotspots, and under the popularity of ChatGPT, "companies will also ask us to watch the AI track." In the Japanese market she is in charge of, she also observed, "Originally, many listed In the second half of the AI company that was dead, a lot of big money came in after the fire."

Like Xiao, William told Ranjiyuan that although he had been paying attention to the AI track before, he mainly focused on face recognition and natural language processing. The AIGC fever brought about by ChatGPT made him start to adjust his focus. Looking at the direction of chatbots, there are also new application scenarios based on AI large models."

Whether it is an investor who has laid out the AI track early, or a latecomer who has just turned his attention, while he is busy digging gold on the AI track that is "beautiful and charming", it is also because of some hot entrepreneurial projects " I have a terrible headache".

"A lot of start-up companies nowadays don't actually have their own technological innovation at all. They just connect to ChatGPT, package the front end, make a UI design, and put it on the Apple store for sale." William said bluntly.

This kind of "pipi" AIGC project lacks technical barriers and mature business logic, and is more just for profit. "This kind of project often only needs to ask a few questions, such as what is your algorithm? What is the technical barrier?" Where? You can test it out." William said.

"Take digital people as an example. Now domestic entrepreneurial projects say they are original developments. In fact, many (projects) are not at all. In the end, they are all fighting for low prices, and the application scenarios are limited. There is not much to see." Previously worked Xiaohongshu blogger @晓昊 from a domestic dual-currency fund told Ranjiyuan, "There is no shortage of funds in the AI track, but many good projects are overseas."

And to identify hot entrepreneurial projects, the more direct method may be to identify the founders.

Wang Xiaohui told Ranjiyuan, "In the past, capital liked entrepreneurs from big factories, but now capital prefers serial entrepreneurs. After all, past experience will not deceive people. In the field of hard technology investment, we pay more attention to whether the founders have the ability to make great achievements. Disruptive technological innovations and industrialization of these technological innovations. In this way, we will pay attention to their scientific research background, whether they have published relevant papers in top international scientific journals, whether they are the main authors of the papers, and whether the team has commercialization ability and so on."

"But in the field of AIGC, we pay more attention to the end-to-end model, and hope that more people familiar with the industry can transform the capabilities of artificial intelligence into efficient tools." Wang Xiaohui added.

Train AI to work for itself

Under the upsurge of AI large-scale model entrepreneurship triggered by ChatGPT, the value of assisting investors in the work of AIGC is more direct and easy to implement.

"For me, ChatGPT is simply the most efficient search portal for research reports (research reports)." William Wong, managing director of Unknown Capital, told Ranjiyuan that as an investor, one of the daily tasks is to search for industry research reports. But in the past, if you want to do in-depth research on an industry, you need to switch multiple browsers and brokerage websites to download more than 10 research reports.

"After searching with ChatGPT, it can automatically grab research reports, which can save me at least one to two days of time." William Wong said excitedly.

"Investors must learn to train ChatGPT to work for themselves." Wang Xiaohui told Ranjiyuan, "With ChatGPT, you can easily capture the key points of dozens or even hundreds of pages of industry research reports and form a mind map."

Wang Xiaohui graduated in 1991 as one of the few majors in economics and management who had access to investment knowledge. In the era without office software, Wang Xiaohui even needed to spend 2-3 weeks to build a financial model. Now with AI tools, it also makes the past The lengthy document preparation time has been shortened a lot.

In addition to using ChatGPT to work for themselves, investors are also using AI to replace part of the work such as design and code writing.

"ChatGPT can replace part of the design manpower." Wang Xiaohui said, because one of the data sources used to train ChatGPT is Github, and Github is the largest software open source community. Some common programs only need to input the required functions, and ChatGPT will It can output basic code, and many software companies are already using ChatGPT to improve the productivity of software developers.

"Using AI to replace repetitive mechanical work can also help us reduce some of our expenses." Wang Xiaohui said.

Like Wang Xiaohui and William, Xiao said that in addition to using AI for document summarization, ChatGPT has become her "personal secretary" in the overseas English communication environment. "ChatGPT can help correct grammar problems, correct expressions, and use Very handy for writing emails.”

"In fact, ChatGPT can also play Soros, Buffett and other roles. Its logic is very strong and can assist decision-making." @晓昊来过 added.

Recently, according to the experimental results released by finder.com, a British financial consulting website, starting from March 6, the net value of the stock portfolio constructed by 38 listed companies recommended by ChatGPT rose by about 4.9% within five weeks, outperforming the UK 10 popular fund products (with an average return of -0.8%) recommended by the online investment platform Interactive Investor, some European and American quantitative investment institutions are also trying to use ChatGPT to build investment strategies.

Figure/ChatGPT’s answer on investment source/provided by Colin

Obviously, the "fishing" attempts of "smart" investors to use ChatGPT to improve work efficiency still need more exploration.

However, @晓昊来过 also bluntly said, "Although AI is already perfect in assisting the creation of standard documents", "there is an obvious positive correlation between the output level and the level of training, patience, and time, but it cannot be ignored." Artificial intelligence, too, has limitations. "

"The biggest problem with ChatGPT is serious nonsense." Regarding this, Wang Xiaohui complained, "Although ChatGPT is used to check research reports, it can really save a lot of time to focus on research reports, but the information he grabs is also very important. Second confirmation is required, and some key information is sometimes ignored.”

Therefore, for Wang Xiaohui, it is not possible to completely rely on AI to replace daily work.

William also said that the use of ChatGPT to search for industry research reports is indeed "fast and comprehensive", but sometimes "AI's search angle may be different from the angle you want." It can only be "guided" to find more by constantly adding questions. Supplementary information, "He can only be trained as an assistant."

Investment boom? Still relatively rational

Capital has the most sensitive sense of trends.

According to the statistics of the investment institution a16z, the annual revenue of AIGC products of image generation, copywriting and code writing has exceeded 100 million US dollars globally; With an investment of one billion US dollars, OpenAI's market value is as high as US$29 billion; Google has invested about US$300 million in Anthropic, which is considered to be a competitor of OpenAI.

According to Bloomberg on March 4, Stability AI (the company behind the Stable Diffusion image generation tool) is seeking to raise funds at a valuation of nearly US$4 billion. The company announced in October 2022 that it had received US$101 million in financing. Jasper.ai, which focuses on AI writing, will raise US$125 million in October 2022, and the company's valuation will increase to US$1.5 billion.

Under the AIGC upsurge, investors should also be the most sensitive group of people. But compared to the global scale and the hot public opinion, is the AIGC craze in the domestic venture capital circle really coming?

According to the data of Qichacha, as of the end of May, a total of 5 related financings in the field of AI text/image/video generation have been completed in 2023, and the publicly disclosed financing amount exceeds 80 million yuan, only involving the two brands of Shenyan Technology and TIAMAT. However, other companies or platforms such as 6pen, draft.art, Domo, Pirates of Dreams, etc. have not yet entered the financing stage.

At the same time, according to the qubit data, as of March 2023, the institutions that have attracted the attention of top capital in my country's AIGC track are all related to AI large models or the end of AI leaders. Among them, Wang Huiwen announced the next The round of financing has been subscribed for 230 million U.S. dollars, Wang Xiaochuan raised 50 million U.S. dollars, etc., and although many institutions have clearly proposed to use AIGC as the main investment track, the main force is still in a wait-and-see state.

In addition to actively looking at the project, William also bluntly said, "The concept of AIGC in China is still in its early stages. We want to see whether there are still entrepreneurs in the market after the trend recedes."

In this regard, Xiao said that although he has seen a lot of AI projects, he has not actually started. "My investment philosophy is to look at the real business value, and I will never invest in whatever is hot."

Whether it is William or Xiao, in the AIGC boom, prudent investors are looking for "good projects that can really empower the real industry."

Xiao told Ranjiyuan that she is now paying attention to an AI company whose AI data model optimizes the traditional industrial measurement process. It is necessary to provide preliminary measurement data, and the first version of the design drawing can be directly generated through algorithm optimization.” For her, this is where the real commercial value lies.

William and Xiao are obviously not the only investors with a cautious attitude. Overseas investment director Ke Lin told Ranjiyuan that although he frequently pays attention to AIGC's entrepreneurial projects, he currently has no investment plans.

"It is difficult to predict the market growth space for domestic AIGC projects, and even new changes appear every day, and many companies are in the stage of storytelling," Ke Lin said. In addition, domestic laws and regulations related to AIGC are not yet perfect. Before the regulatory "boots" are implemented, the AIGC industry is obviously still highly uncertain.

However, according to qubit estimates, my country’s AIGC track will usher in an exponential growth in financing in 2023, and the AIGC market will exceed one trillion yuan in 2030. Looking at the global market, according to market research company Brainy Insights, the generation AI The global market will grow from US$8.65 billion in 2022 to US$188.62 billion in 2032, with an average annual growth rate of 36.1%.

Under the upsurge, it is naturally a good choice to board the express train of AIGC first. But investors who use money to choose must also be more cautious when choosing projects. AIGC's investment boom may not yet come.

Reference materials:

"China AIGC Industry Panoramic Report and AIGC+50 List", source: Qubit.

*Colin and Xiao are pseudonyms in the text.