Artificial intelligence companies have entered the stage of stock game.

Recently, Intel announced the fifth phase of the list of innovation teams of the AI ​​Top 100 Innovation Incentive Program, and jointly released the “China Growth AI Enterprise Research Report” with Deloitte and the Shenzhen Artificial Intelligence Industry Association.

Li Desheng, managing director of Intel’s China Strategic Cooperation and Innovation Business Department and head of Intel’s Innovation Accelerator, mentioned that since the establishment of the AI ​​Top 100 Innovation Incentive Program in 2018, 78 outstanding AI innovation companies have joined. The 16 innovation teams in the new phase are selected from more than 300 AI innovation companies, with a total valuation of more than 26 billion yuan, including: Nebula clustar, Impulse Online, Aishu Smart, Rongxu Smart, Real Smart, Yi Daoboshi, Yushu Technology, Deepzhi Technology, Laiye Technology, Darerguan Technology, Blue Fatty Machine Intelligence, Huaqing Kesheng, Shaanxi Vision Intelligent Manufacturing, Coopert Technology, Puzhou Aircraft Technology, Weixin Ingenuity.

What is the use of large enterprise innovation accelerators?

In recent years, global technology giants have launched an endless stream of incubators, but everyone has mixed opinions on the value of incubators. The “Research Report on China’s Growth AI Companies” released this time explains this issue:

First, the incubator will provide new national policy information to start-up companies in a timely manner, and will accurately push the key support companies. Taking Zhongguancun Chuangye Street as an example, the state provides financial support of no more than 10 million yuan per year for the purchase of professional equipment and materials, operating expenses, and house leasing for companies that focus on hard technology innovation, and continuous support for no more than 3 years;

Secondly, incubators can bring phased and diversified help to start-ups, especially those that develop innovative products. For example, in the start-up stage of a company, the most important thing is to find the direction of development, including what problems the company has to solve and how to design the equity structure. For growing AI companies at this stage, incubators can help them set up companies for free, and help start-ups with low-cost service products to do a good job in the early operation of the company, so that start-ups can better focus on product research and development; launch products in the company In the later stage of growth, companies need to do a good job of financial planning, exposure and publicity, so as to obtain more market resources. At this stage, the incubator will accurately push relevant government support policies, assist companies in market promotion and find customers, and connect with capital market resources.

As a company in the second phase of the “AI Top 100 Innovation Incentive Program”, Liu Ruoshui, partner of Extreme Vision Technology, said that Intel has given them a lot of support in terms of technology and brand marketing.

First of all, at the technical level, Liu Ruoshui mentioned: “When joining the AI ​​Baijia Innovation Incentive Program, every company will choose a project that is being done to start support. At that time, the project selected by the extreme perspective was the joint venture with the Ministry of Public Security. Institute, Traffic Management Technology Research Institute OneThe algorithm of the traffic violation review. At that time, the server resources needed for the algorithm platform were relatively expensive, but Intel’s system provided us with technical support, allowing us to adapt to better performance and cost-effective under the condition of ensuring the accuracy of the algorithm. A hardware server, so that the efficiency of the entire algorithm can be guaranteed, and the cost that the customer needs to bear when the application is implemented is also controllable. In addition, before cooperating with Intel, we don’t know much about hardware, and the algorithm application scenarios are very focused. It is more complicated to choose a suitable server. However, through the AI ​​Baijia Innovation Incentive Program, we picked the best one. The server that adapts to the scene is equivalent to forming a standard product, which is a standard product of software and hardware in our algorithm mall. The landing process is very convenient. Our set of products is the first AI traffic police product certified by the Ministry of Public Security in the country. It has been launched in 13 cities. Traffic police brigades in Guangxi and Jiangxi are already using this product. “

Secondly, at the level of brand promotion, Liu Ruoshui mentioned: “Our project happened to land during the epidemic. Because of the epidemic, it was delayed for more than a month, but it did not have much impact on the market. We are in This product was completed around the middle of the year. Intel used its strong market capabilities to re-package our products and help us to promote them in Intel’s channels. The follow-up will be very fast. Fifty integrators contacted us, hoping to apply this product.”

The status quo of growing AI companies

The “Research Report on China’s Growth AI Companies” released this time stated that the current Chinese AI-related companies can be divided into four categories: Internet giants, traditional software and hardware vendors (including integrators SI, independent software developers ISV , Equipment manufacturers OEMs, leading companies in various vertical industries, etc.), AI unicorn companies, AI growth companies.

Among them, there are about 4,484 AI growth companies, accounting for 89% of AI-related companies, and the number is huge. However, the number of newly established companies has decreased in recent years, from 528 newly established in 2016 to only in 2020. 25 newly established companies. Artificial intelligence companies have entered the stock game stage. In the next few years, the entrepreneurial requirements for entering the AI ​​market will be higher and the difficulty will increase.

In addition, in 2020, there will be 34 artificial intelligence unicorn companies in China, of which 10 unicorn companies are valued at over 10 billion yuan. But behind the prosperity, unicorn companies generally have low profit margins or even losses. In the face of the squeeze of the same technology platform-based Internet giants and the increasing labor costs, the money-burning model is difficult to sustain for a long time. In addition, the commercialization of technologies and products is a common aspect of unicorn companies.Pro bottleneck.

Overall, in 2020, China’s investment and financing in the field of artificial intelligence will reach a new high, reaching 174.8 billion yuan, a year-on-year increase of 73.8% compared to 2019, but the number of financings has decreased year-on-year, average single financing The amount has steadily increased, from an average single financing of 66 million yuan in 2016 to an average of 266 million yuan in 2020. On the whole, the amount of investment and financing before the A round has been significantly reduced, funds continue to be concentrated in the B round and subsequent rounds of enterprises.

Among them, smart manufacturing and smart healthcare are the two sub-sectors with the largest number of investment and financing. In 2020, the largest single domestic artificial intelligence financing appeared in the medical field. Minimally Invasive Medical completed a number of first difficult surgical operations with endoscopic surgical robots, and won a single financing of RMB 3 billion. Yikatong Technology has received 1.3 billion yuan of RMB A round of financing by virtue of high-performance automotive-grade digital cockpit chips. Jifei Technology received 1.2 billion yuan in Series D financing led by Baidu Capital and the second phase of the Softbank Vision Fund with the integrated software and hardware solutions for unmanned farms.

Let some people see the future first