digital china 2029 | the future is not yet in our language, can ai create it?
2024-09-13
한어Русский языкEnglishFrançaisIndonesianSanskrit日本語DeutschPortuguêsΕλληνικάespañolItalianoSuomalainenLatina
"artificial intelligence technology is developing rapidly, with the iteration cycle shortened from 'years' to 'weeks'. the 'emergence' of innovative artificial intelligence technologies is basically one wave after another."
zhu jiaming, a well-known economist and chairman of the academic and technical committee of the digital chain digital finance research institute of the hengqin guangdong-macao deep cooperation zone, expressed the above views at the "digital china 2029: embracing the new generation of information technology" seminar held by the pengpai research institute on september 6.
in december 2022, the generative artificial intelligence chatgpt was born, bringing the world into the era of artificial intelligence. the 2024 government work report first proposed the "artificial intelligence +" action to promote the deep integration of artificial intelligence with all walks of life.
the digital economy is reshaping the world economic landscape at an unprecedented speed and scale. the rise of a new generation of information technology is becoming a key driving force for china to achieve high-quality economic development. against this background, sorting out the development of artificial intelligence, interpreting its applications and bottlenecks, and looking forward to the future civilization order are not only of practical significance for the present, but will also provide inspiration for the future.
on the afternoon of september 6, the paper research institute held a seminar on "digital china 2029: embracing the next generation of information technology". four guests, zhu jiaming, shi dan, peng wensheng and zheng lei, gave keynote speeches. the picture shows the roundtable discussion.photo by the paper reporter zhou pinglang
general technological changes and digital factors have changed the rules of economic operation
globally, mankind is entering the era of the fourth industrial revolution, the digital revolution. shi dan, former director and researcher of the institute of industrial economics of the chinese academy of social sciences, pointed out at the seminar that the digital revolution is different from the previous three industrial revolutions. its impact is not limited to a specific industry or technological breakthrough, but in the form of a general technology, it has profoundly changed the way all industries operate.
the core of this change is that data has become a production factor as important as capital and labor. the first three industrial revolutions were dominated by a certain industry and drove economic development. the digital revolution, on the other hand, has promoted the digital transformation of the entire economic system and has had a huge impact on every industry.
the uniqueness of the digital revolution lies in its all-round application as a general technology. digital technology has not only promoted the vigorous development of the digital industry, but also become an indispensable element in the production process. unlike traditional production factors such as natural resources such as oil, coal or natural gas, digital factors have the characteristics of non-exclusivity, non-depletion and increasing returns to scale. therefore, digital factors can be used simultaneously in multiple production processes, will not be exhausted, and their value increases with the increase of application scenarios.
shi dan pointed out that with the development of the digital economy, the role of digital factors in the production process has become increasingly prominent, which not only helps to expand production, save resources and reduce costs, but also can give birth to new industries, new models and new business formats, thereby fundamentally changing the traditional economic activities, economic operation rules and governance methods.
therefore, industrial development in the digital era presents new characteristics. shi dan summarized it into four trends: first, the rise of digital industries and the acceleration of industrial digitalization. digital industries make economic growth no longer dependent on the input of natural resource elements, but drive economic growth with new technological innovations. shi dan said: "now, the annual average growth rate of the global digital industry has reached about 20%, becoming an important force driving world economic growth." at the same time, digital transformation has become a key driving force for industrial upgrading. through the digitalization process, enterprises have not only improved the efficiency of internal resource allocation, but also promoted the efficient circulation of the entire industrial chain.
secondly, integrated development has gradually become a new driving force for growth. the traditional division of labor model has been broken, and industrial integration, product integration and market integration have become new drivers of economic development. shi dan pointed out that the in-depth development of the integration of digital industries and traditional industries has spawned many new formats and models, changing the correlation between industries.
third, the importance of coordinated development of the industrial chain and platform enterprises is becoming increasingly prominent. the industrial organization mode is evolving into a more networked and platform-based division of labor system. platform enterprises therefore play a vital role, which can efficiently connect market supply and demand, determine market prices, and grow very fast. among the top ten fortune 500 companies, eight are platform-based companies.
fourth, the coordinated transformation of digitalization and greening has become an important issue. promoting the digital economy and green and low-carbon development requirements have become a prerequisite. the high energy consumption of digital technology has led to a substantial increase in energy demand. therefore, future industrial transformation must develop in the direction of renewable energy, build a new energy system, and achieve the coordinated development of digitalization and greening.
the bottlenecks in public data opening still need to be solved
digital factors play an important role in the institutional construction of the deep integration of digital and real economy and the international competition of large-scale artificial intelligence models. it is urgent to make up for the shortcomings in data.
jiang xiaojuan, a professor at the university of chinese academy of social sciences, said at the digital wilderness symposium in may this year that china's advantages are a strong public sector, good data volume and data structure. the government should vigorously promote china's institutional strengths as soon as possible.
however, she also pointed out that research shows that the data resources controlled by the chinese government account for more than 50% of the total data resources in society, but they are not open to sharing enough and the efficiency of data utilization is not high enough.
zheng lei, professor at the school of international relations and public affairs of fudan university and director of the digital and mobile governance laboratory, said at the meeting that public data opening faces two bottlenecks. first, high-quality data cannot be "disclosed". on the one hand, the government is worried about security issues, and data opening may involve security, privacy, business intelligence and other issues; on the other hand, the government is not willing enough, and providing data free of charge increases the workload and burden.
the second bottleneck is "unable to flow". in the middle link of supply and demand matching, the government as the supplier does not know what kind of data the market needs, and the enterprises as the demanders do not know what kind of data the government has. in addition, the government's technical development and operation service capabilities are limited, making it difficult to match supply and demand. the government's lack of manpower and financial resources also makes it impossible to continuously and stably supply high-value data that needs to be processed.
to solve the bottleneck, "authorized operation" is a model being explored. the main difference between data openness and authorized operation lies in the operation mode and applicable scenarios. data openness emphasizes the free and extensive provision of data to promote innovation and enhance public interests; while authorized operation focuses on protecting data security and privacy, and authorizing third-party entities to process raw data into services or products for use by users in specific markets. this model is usually provided based on scenario orientation, that is, "one scenario, one approval", and requires the operating entity to play an intermediary role.
however, authorized operations also face many challenges, especially in the distribution of "rights, responsibilities and benefits". for example, the responsible departments are still not willing enough; which entities should be involved in the distribution of benefits. in addition, not only public data, but also platform data has "ownership" issues. for example, the records of consumers on shopping platforms belong to both the platform and the consumers.
zheng lei said: "we should focus on making data flow while protecting the rights and interests of all parties, and distribute the benefits to everyone, rather than determining its ownership."
how should we imagine ai now that embodied intelligence is in the baby boom era?
in december 2022, the american artificial intelligence company openai launched the generative artificial intelligence chatgpt, which became a disruptive innovation in the history of artificial intelligence development.
in the 20 months since then, ai big models have developed unprecedentedly, bringing ai into the stage of generative ai. "in my opinion, in these 20 months, the actual development of ai has been based on a time scale of two weeks, or at the latest one month." zhu jiaming believes that the emergence of agentic ai marks the entry into the "baby boom" era of embodied intelligence. "it is now growing at a rate of tens of thousands or hundreds of thousands, and i believe it will become the main body that affects the economy in three to five years."
peng wensheng, chief economist of cicc, pointed out that china will have significant advantages in the field of embodied intelligence in the future. embodied intelligence requires highly flexible and precise mechanical movement capabilities, and china's manufacturing infrastructure and technological accumulation have given it unique competitiveness in this cutting-edge field.
zhu jiaming believes that artificial intelligence will change the real economy mainly through two modes. in the first mode, ai development itself has industrial genes, and each step of its advancement naturally leads to the emergence and development of new industries. there are many cases of this mode. openai is a new company related to large models. it is itself a new industry. it has industrial technology, direction, business model and typical products.
the second model is ai transforming traditional industries, sectors and departments. zhu jiaming said: "no industry can reject artificial intelligence, because the greatest contribution of artificial intelligence is to ensure that the labor productivity of traditional industries can be improved, and the computing power and algorithms of any industry, enterprise and economic organization can be improved."
similarly, peng wensheng also mentioned "industry ai" at the meeting. he believes that china occupies an important position in the global manufacturing supply chain, especially in the leading position in the fields of robots, photovoltaics, air conditioning, new energy vehicles, etc., which enables it to realize the intelligent upgrade of production processes through ai technology. this upgrade can not only improve production efficiency, but also reduce costs, further enhancing the global competitiveness of china's manufacturing industry.
zhu jiaming has nine observations on how ai will affect the future economy. first, the main body of economic activities has changed from natural persons to the participation of intelligent entities. second, the change of economic entities has led to changes in people themselves, and humans are becoming more and more intelligent. humans are surrounded by various artificial intelligence tools, various wearable devices and various brain interfaces. third, the space and time of human activities have changed. fourth, the economic cycle has changed. fifth, the product form has changed. from high-tech content to high-intelligence content. sixth, the economic distribution system has changed. humans have greatly reduced working hours and work intensity, and the government and the state have guaranteed the basic income of the people, which has become a basic feature of the artificial intelligence era. seventh, the basic structure of economic development has been changed and adjusted. infrastructure will be participated by artificial intelligence. eighth, the relationship between production and consumption has changed. in the supply chain of the artificial intelligence era, many productions are consumption, and many consumptions are production. ninth, the world pattern will be changed. the level of development of the artificial intelligence industry will determine the relationship between countries.
zhu jiaming further pointed out, "now the real economy itself has begun to digitize, and all digital economies have been integrated with the real economy. the core is still to solve the problems of computing power and algorithms."
as for the development stage of artificial intelligence and their imagination of the future, experts had a clash of views at the meeting. peng wensheng believed that artificial intelligence has crossed the first inflection point of the s-shaped curve and entered a stage of rapid growth. the key to the next step of development lies in how to smoothly realize the industrial application of ai and the ai transformation of the industry.
zhu jiaming expressed reservations about whether ai has reached the turning point of full application. "artificial intelligence is a process of continuous innovation, where one innovation drives another. there is no stable turning point in the process, after which artificial intelligence shifts from research to application." "today, every advancement in artificial intelligence becomes a prerequisite for application, and every advancement must be guided by research, otherwise it will be the beginning of lag and backwardness when it comes to application."
at the same time, zhu jiaming's imagination of what new forms of artificial intelligence will develop in the future is even bolder. "the intelligent agent derived from the artificial intelligence generative model is no longer an artificial intelligence tool that serves people and is passively driven by people. the biggest feature of the intelligent agent is that it is proactive, automatic, and adaptable. it can become an intelligent form that is not affected or controlled by people."
stan said: "as to whether artificial intelligence can operate independently of humans in the future, some people have proposed the idea of 'new silicon-based organisms'. but i think this discussion is still open, and i personally do not agree with this statement. according to the marxist view, society is still centered on people."
peng wensheng is also relatively conservative. "because the big model is based on existing knowledge and carriers, whether it is video, text, or pictures. the knowledge that humans don't have yet and the knowledge that humans will innovate in the future do not exist in our language or in our pictures. how can we use the existing language to find patterns and discover new things? i personally don't believe that ai will have this ability to think and innovate in the future. of course, i may be wrong. this issue is very controversial."
consumption or green, unemployment or security, what will ai bring to the future?
it may be unknown whether artificial intelligence will bring risks and challenges beyond human control in the future. however, some impacts can already be foreseen.
the cicc research team conducted a quantitative analysis of the development potential of ai from the perspectives of technology activity and market activity, and estimated the ai development index. under this analytical framework, the united states leads with its technology activity, while china shows great potential with its market friendliness.
"ai is expected to increase china's gdp by 9.8% in 2035 compared to the baseline scenario, equivalent to an additional 0.8 percentage points in the annualized growth rate over the next 10 years." peng wensheng said that china's huge population size and diverse application scenarios provide ideal conditions for the large-scale application of ai, which may bring huge dividends similar to those in the internet era.
however, the economies of scale brought by artificial intelligence also face many challenges. peng wensheng said that many factors, including energy consumption, data governance, ethics and security, may limit the full realization of ai's economies of scale.
shi dan said that green and low-carbon development must be achieved in the process of promoting the digital transformation of the industry. in theory, the kuznets curve (ekc) shows that environmental improvement can be achieved with economic development. but shi dan pointed out that considering the situation of climate change, the kuznets curve (ekc) is actually not valid. some scientists believe that if the temperature rises by more than 2 degrees celsius, the atmospheric changes are irreversible and the environment cannot be repaired. "this is different from the simple pollution of wastewater and waste gas emissions that we talked about before." therefore, green development must be taken as a prerequisite for digital transformation, or to promote coordinated development.
in addition to climate, another topic that has sparked widespread discussion is whether the development of artificial intelligence will lead to large-scale unemployment.
in this regard, the research results of cicc's research team show that the impact of ai is more significant in the field of physical labor, especially in the mining industry, resource processing industry and other labor-intensive tasks, which have the largest replacement space. in the next decade, ai is expected to significantly improve the productivity of these industries, but the productivity improvement in industries such as wholesale and retail, accommodation and catering will be relatively low. "therefore, we believe that the popularization of ai is more likely to drive the labor force to shift to more creative and valuable work areas, thereby optimizing the overall employment structure, rather than causing large-scale unemployment."
zheng lei believes that if the intelligent era may have an impact on the real economy and employed people, then the government needs to adjust its policies on labor security and social distribution. in this regard, peng wensheng believes that the development of artificial intelligence is expected to enhance the overall ability of society to improve social security. with the economic growth and improved production efficiency driven by ai, it is expected that china's social security system will usher in major improvements in the next 5 to 10 years.
(intern cui yiming also contributed to this article)
shao yuanyuan, researcher at the paper
(this article is from the paper. for more original information, please download the "the paper" app)
