news

In the field of artificial intelligence, the investment boom seems to be experiencing a cooling-off period.

Although AI technology is widely regarded as a key driving force for the future, the reality is that many AI startups do not solve the real pain points of the market, resulting in a mismatch between funding supply and project demand. Investors are beginning to more rationally evaluate the market potential and practical application value of AI projects. At the same time, large-scale model projects are more likely to be favored by capital due to their technical barriers and broad application prospects.

However, even these projects face challenges such as uncertainty in technological development, high R&D costs, and difficulty in commercialization.

What is the future of AI investment?

“In the AI ​​startup sector, the supply of projects seems to exceed demand,” said an ALL-in AI investor.

The ideal AI startup process should be to develop solutions based on a specific market demand, and then show investors the breadth and urgency of this demand. However, in reality, many projects do the opposite. They first build a multi-dimensional artificial intelligence platform, and then look for funds and determine the specific direction. This approach makes it difficult for investors to evaluate the specific profit model and market potential of the project.

From the perspective of capital supply, although US dollar funds are very interested in AI startups, they have also encountered difficulties in fundraising this year, which has limited their investment in the high-risk, long-term AI field. At the same time, some mature RMB funds are also paying attention to the AI ​​field, but they prefer to invest in projects that have been verified by the market and have achieved certain results.

Take autonomous driving as an example. This field was once seen as one of the most promising applications of artificial intelligence. However, there is a gap between reality and ideal. Autonomous driving requires not only powerful technology, but also the support of policies and regulations and the cooperation of market promotion. For example, the road rules for autonomous driving vehicles and the definition of accident liability need to be gradually improved. In addition, automobile manufacturers' publicity to consumers is also the key to improving the acceptance of autonomous driving. These factors have prolonged the process of commercialization of autonomous driving and affected the enthusiasm of investors. From the beginning of the autonomous driving boom in 2016 to 2021, the number and scale of investment and financing events in the domestic autonomous driving industry have declined, and they have decreased significantly in 2022 and 2023.

Although AI has innovative projects in many fields such as office, creation, and education, many investors who have been deeply involved in the hard technology field for many years believe that there has not been enough eye-catching innovation scenes this year. They believe that the AI+ model needs to have disruptive innovations that can truly improve production efficiency, rather than just superficial decoration.

The investment situation in the field of AI has shown a certain differentiation. Large model projects are relatively easy to obtain funds, while AI projects targeting specific application scenarios are more difficult to finance. An AI investor pointed out that the main reason is that these application projects have failed to effectively solve practical problems. For example, an essay correction project developed by an entrepreneur in Chongqing has been favored by investors because it solves practical needs. Some people even chased to the airport to intercept it for investment. Similarly, the AI ​​medical companion service provides parents with rapid diagnosis and medical decision-making support when their children are sick at night, which is also seen as solving practical pain points.

In contrast, projects that rely solely on information asymmetry, teaching or AI entrepreneurship training, or simply add technologies such as AI face-changing and generative content to existing projects, are less likely to obtain financing.

Last June, an AI collaborative office project developed by a doctoral team in Singapore received financial support from individual investors. Xiaoyang (pseudonym), one of the team members, said that AI investors are cautious about the current craze. If there is no technical barrier, only a conceptual business plan, no tested products, or the product is easy to copy, then the possibility of obtaining financing is almost zero.

Xiaoyang also introduced that the core members of their team are all from the doctoral team of the University of Singapore, and they have been involved in the language model project for three years. In the first half of this year, the project developed by the team has passed the test, and the marketing staff is establishing commercialization channels with B-side enterprises. Since the team members are all doctoral students, they often participate in the school's innovation competition and get to know some investor resources through their mentors. One of the individual investors showed great interest in their project and invested when the project was two-thirds completed, mainly to cover R&D expenses.

Xiaoyang pointed out that AI has attracted attention because consumers see its potential in more application scenarios. However, for entrepreneurs and investors who have been in this field for a long time, this is not a new or sustainable growth point. Therefore, when most investors receive business plans under this wave of popularity, they will rationally examine the team background and the time they have been in the industry, and they will be relatively conservative in the investment amount, usually only covering the next stage of R&D expenses, and then make additional investments after progress is made.

In recent years, a number of AI unicorn companies have emerged, including Zhipu AI, Baichuan Intelligence, and Zero One Everything. In the subdivided fields, computer vision and imaging at the technical level, intelligent robot projects at the application level, and intelligent driving/autonomous driving projects are the areas that have received more venture capital.

When it comes to AI companies that have been highly sought after by investors in the past two years, there are some commonalities. Big models are naturally one of the favorite AI investment directions of investors. Many AI unicorn companies focus on developing and utilizing big model technologies, such as Zhipu AI and Zero One Everything. They have launched various applications based on big models, including but not limited to natural language processing, image and video generation, etc.

In addition, some companies have also developed cross-modal models of text to visual, text to audio, and text to text, demonstrating AI's ability to process different types of data.

Among AI unicorn companies, open source collaboration is often the method they often choose. For example, Zhipu AI has open-sourced the Chinese-English bilingual dialogue model ChatGLM-6B, which has promoted technology sharing and collaboration and accelerated the development of AI technology. Some unicorn companies focus on applying AI technology to specific industries or vertical scenarios, such as Zhiyuan Robotics' research and development in the field of humanoid robots and Meijia Technology's innovation in the field of automotive intelligence and networking components.

In addition, many AI unicorn companies have founding teams from top technology companies or well-known universities. These team members often have rich research and practical experience, providing a solid foundation for the company's technological innovation.

Even though they are highly sought after, the biggest challenges faced by AI unicorn companies in the process of technological innovation mainly come from the uncertainty of technological development, the high cost of R&D, and the difficulty of commercializing technology.

Technological development is changing with each passing day, and AI companies need to keep up with the latest technological trends. This not only requires companies to have continuous R&D investment, but also requires R&D teams to be able to quickly adapt to technological changes and maintain innovative vitality.

High R&D costs are another major challenge facing AI unicorns. Especially in the field of large models and deep learning, training and deploying these models requires huge computing resources and data support, and these resources often require expensive hardware investment and operation and maintenance costs. In addition, the commercialization of AI technology is also full of challenges. How to transform technology into products accepted by the market and how to find a suitable business model and profit path are all problems that AI unicorns need to solve.

Data privacy and compliance risks are also challenges that AI unicorn companies cannot ignore in the process of technological innovation. As data protection regulations become increasingly stringent, companies must ensure that their data processing and application comply with the requirements of laws and regulations, which not only increases compliance costs, but also places higher demands on the design and application of technology.

The stringent investment requirements have also made investors gradually cooler about the AI ​​field.

The financing situation of China's artificial intelligence industry in the primary market in 2023 shows some changes. Although the overall investment and financing industry has cooled down, the financing situation in the field of artificial intelligence is still relatively hot. According to IT Orange data, as of November 20, 2023, the total number of financing events in the primary market of the artificial intelligence track was 530, a decrease of 26% compared with the same period last year; the total financing transaction amount was estimated to be 63.1 billion yuan, a decrease of 38% compared with the same period last year.

The AI ​​investment field is experiencing many challenges and difficulties.

As technology continues to advance, investors are paying more and more attention to the actual return on investment (ROI) of AI projects. Although the return on investment of AI projects has increased in recent years, the market's expectations for the benefits of AI investment have shifted from pure technical capabilities to actual benefits. In addition, the rapid iteration of AI technology requires companies to continue to follow up on the latest technology, which means that investors need to have a deep understanding and accurate judgment of technology trends.

AI projects, especially the training of large models, require huge and continuous R&D investment, which is a huge challenge for startups with limited funds. At the same time, AI companies need to find scalable implementation scenarios during the commercialization process to transform technology into actual business value, a process full of uncertainty. Data privacy and compliance risks have increased with the implementation of the Data Security Law and the Personal Information Protection Law, bringing higher compliance costs and risks to AI companies.

The market's enthusiasm for AI investment has also triggered concerns about financial bubbles. Some analysts believe that the current huge investment in AI may lead to a financial bubble because the gap between AI benefits and expenditures is widening, and huge revenues are needed to ensure that the investment produces the expected return. In addition, although AI technology is developing, there is a lack of AI products that bring substantial benefits in the market, and there are not many AI products that consumers actually use.

In the investment field, the valuations of some AI companies may be too high, making it difficult for the primary market to take over, and in the case of continued losses, the secondary market may not be willing to pay. As the leading AI companies' ability to attract funds increases, single large-scale financing events have occurred frequently, which not only raises the overall average financing amount of the industry, but also increases the threshold for investment. The uncertainty of the policy and regulatory environment is also a factor that investors need to consider. Policy changes may affect the operations and investment returns of AI companies, especially in terms of data use and privacy protection.

When considering AI investment, investors need to consider these factors comprehensively and make wise decisions. The complexity and uncertainty of AI investment require investors to have in-depth technical insights, market analysis capabilities, and risk assessment capabilities.

In the field of AI investment, investors face many challenges, the biggest of which include the uncertainty of technological development, the difficulty of commercialization, high R&D costs, data privacy and compliance risks, concerns about market bubbles, and overvaluation. In the face of these challenges, investors need to adopt a series of strategies to assess and manage risks.

First, investors must have a deep understanding of the latest developments and future trends in AI technology, including assessing the maturity of the technology and its market application potential. Understanding technology trends is essential for predicting the direction of the industry. Second, it is key to evaluate the business model of AI companies, especially the market demand for their products or services, customer base, revenue sources, and profitability.

Considering that AI projects usually require huge R&D investment, investors need to carefully evaluate the company's financial status and financing capabilities, as well as the expected return on investment. At the same time, data compliance review is also essential to ensure that the data collection, processing and storage methods of AI companies comply with data protection regulations and privacy policies.

Identifying market bubble risks is crucial to avoid investing at market peaks. Investors should conduct a reasonable analysis of the valuations of AI companies and avoid investing in overvalued projects. In addition, diversifying the portfolio to spread the risk, including investing in AI companies in different technology fields and at different stages of development, can reduce the risk of a single investment.

Continuous monitoring after investment and timely adjustment of investment strategies are essential to cope with market changes. Establishing a risk management mechanism, including risk identification, assessment, monitoring and mitigation measures, can help investors better control potential losses. When necessary, consulting industry experts or professional investment consultants can provide deeper market insights and investment advice.

In short, AI investment requires investors to take a comprehensive and cautious approach to assess risks and ensure that investment decisions are based on sufficient information and professional analysis. Through these strategies, investors can better cope with the challenges in the field of AI investment and achieve a solid return on investment.