In the current context of the rapid integration of the digital and physical worlds, Artificial Intelligence Generated Content (AIGC) is quietly leading a profound transformation, reshaping and even disrupting the production and consumption patterns of digital content, which will greatly enrich people’s digital lives and is an indispensable supporting force for the future’s comprehensive transition to a new era of digital civilization.
Development History and Concept of Artificial Intelligence Generated Content
In 1950, Alan Turing proposed the famous “Turing Test” in his paper “Computing Machinery and Intelligence,” providing a method to determine whether a machine possesses “intelligence,” that is, whether it can mimic human thought processes to “generate” content and interact with people.
In a way, since then, artificial intelligence has been expected to be used for content creation. After more than half a century of development, with the rapid accumulation of data, the enhancement of computational power, and the improvement of algorithms, today’s artificial intelligence can not only interact with humans but also engage in creative tasks such as writing, composing music, painting, and video production.
In 2018, an artwork generated by artificial intelligence was sold at Christie’s for $432,500, becoming the first AI-generated artwork to be sold, attracting widespread attention.
As artificial intelligence is increasingly applied to content creation, the concept of Artificial Intelligence Generated Content (AIGC) has quietly emerged. 
AIGC Historical Evolution
Based on the evolution of artificial intelligence, the development of AIGC can be roughly divided into three stages: the early germination stage (from the 1950s to the mid-1990s), the accumulation stage (from the mid-1990s to the mid-2010s), and the rapid development stage (from the mid-2010s to the present).
Since 2014, with the introduction and iterative updates of deep learning algorithms represented by Generative Adversarial Networks (GAN), AIGC has entered a new era, with generated content flourishing and becoming increasingly realistic to the point where it is difficult for humans to distinguish.
In 2017, Microsoft’s AI girl “Xiaoice” launched the world’s first poetry collection entirely created by artificial intelligence, titled “The Sun Lost the Glass Window.”
In 2018, NVIDIA released the StyleGAN model, which can automatically generate images, and it has now been upgraded to the fourth generation model, StyleGAN-XL, whose high-resolution images are difficult for the human eye to distinguish from real ones.
In 2019, DeepMind released the DVD-GAN model to generate continuous videos, performing exceptionally well in clear scenes such as grass and squares.
In 2021, OpenAI launched DALL-E and released an upgraded version, DALL-E-2, a year later, mainly applied to the interactive generation of text and images, allowing users to input brief descriptive text for DALL-E-2 to create corresponding high-quality artworks in cartoon, realistic, and abstract styles.
The Concept and Connotation of AIGC
Currently, there is no unified definition for the concept of AIGC. The understanding of AIGC in China’s industry, academia, and research circles is that it is a new type of production method that automatically generates content using artificial intelligence technology after Professional Generated Content (PGC) and User Generated Content (UGC).
Internationally, the corresponding term is “AI-generated Media” or “Synthetic Media,” defined as a general term for producing, manipulating, and modifying data or media through artificial intelligence algorithms.
In summary, we believe that AIGC is a type of content classified from the perspective of content producers, a method of content production, and a collection of technologies used for automated content generation.
From the perspective of development background, the rise of AIGC stems from rapid breakthroughs in deep learning technology and the growing demand for digital content supply.
On the one hand, technological advancements drive the increasing availability of AIGC.
On the other hand, massive demand propels the application of AIGC. As the integration of the digital economy and the real economy deepens, and with platforms like Meta, Microsoft, and ByteDance transitioning their digital scenarios to the metaverse, the overall demand for the quantity and richness of digital content continues to rise. Technological capability
In terms of technological capability, AIGC can be divided into three levels based on its target audience and functional implementation.
First, intelligent digital content twins.
The main goal is to establish a mapping from the real world to the digital world, efficiently and perceptibly transferring physical attributes (such as size, texture, color, etc.) and social attributes (such as subject behavior, subject relationships, etc.) from the real world.
Second, intelligent digital content editing.
The primary purpose is to establish bidirectional interaction between the digital world and the real world. Based on digital content twins, this allows for control and modification of content in the virtual digital world from the real world while utilizing the high-efficiency simulation and low-cost trial-and-error advantages of the digital world to provide rapid iteration capabilities for real-world applications.
Third, intelligent digital content creation.
The main goal is to enable artificial intelligence algorithms to possess content creation and self-evolution capabilities, resulting in AIGC products with creative abilities similar to or exceeding those of humans. These three levels of capability together form the closed loop of AIGC’s capabilities. Application value
In terms of application value, AIGC is expected to become a new engine for the innovative development of digital content, injecting new momentum into the development of the digital economy.
On the one hand, AIGC can undertake basic mechanical labor such as information mining, material retrieval, and replica editing with manufacturing capabilities and knowledge levels superior to those of humans, achieving low marginal cost and high efficiency to meet massive personalized demands from a technological perspective; at the same time, it can innovate the processes and paradigms of content production, providing possibilities for more imaginative content and more diverse dissemination methods, promoting content production towards a more creative direction.
On the other hand, AIGC can support multidimensional interaction and integration of digital content with other industries, thereby nurturing new business forms and models, and creating new growth points for economic development, providing new momentum for the development of various industries.
Since 2021, the “metaverse” has shown explosive growth beyond imagination; as the “ultimate” digital carrier of the integration of the digital and physical worlds, the metaverse will feature continuity, real-time interaction, and creativity, and will accelerate the replication of the physical world and infinite content creation through AIGC, achieving spontaneous organic growth.
The Technical System and Evolution Directions of Artificial Intelligence Generated Content
AIGC Technology Upgrades Entering a Deepening Stage
The continuous iteration of artificial intelligence algorithms is the driving force behind the development and progress of AIGC. From the perspective of technological evolution, AIGC technology can be roughly divided into the traditional template or rule-based pre-deep learning stage and the rapidly developing deep learning stage of deep neural networks.
The Potential of AIGC Large Model Architecture is Highlighted
The rapid development of super deep learning in recent years has led to continuous breakthroughs in deep neural network technology in both large models and multimodal approaches, providing strong support and new possibilities for upgrading AIGC’s technological capabilities.
AIGC Technology Evolves Three Major Frontier Capabilities
AIGC technology is widely applied to audio, text, visual, and other different modal data, forming a rich variety of technological applications. These include intelligent digital content twin capabilities, intelligent digital content editing capabilities, and intelligent digital content creation capabilities.
In addition to repairing and enhancing content of various modal data, intelligent enhancement technologies in the field of three-dimensional vision have made rapid progress in recent years.
Based on digital content twin technology, intelligent digital content editing technologies have constructed interactive channels between the virtual digital world and the real physical world.
From a technical perspective, intelligent digital content editing primarily achieves content modification and control through two types of technologies: semantic understanding of digital content and attribute control.
Source: JD Exploration Research Institute
Application Scenarios of Artificial Intelligence Generated Content
1. AIGC + Media: Human-Machine Collaborative Production, Promoting Media Integration.
In the editing phase, first, it realizes voice transcription of interview recordings, enhancing the work experience of media workers. During the 2022 Winter Olympics, iFlytek’s intelligent recording pen assisted reporters in quickly producing articles within 2 minutes through cross-language voice transcription.
Second, it enables intelligent news writing, improving the timeliness of news information.
Third, it achieves intelligent video editing, enhancing the value of video content. During the 2022 Winter Olympics, CCTV used AI intelligent content production and editing systems to efficiently produce and release video highlights of winter sports, creating more possibilities for deeply developing the value of sports media copyright content.
In the dissemination phase, AIGC applications mainly focus on AI-generated anchors for news broadcasting.
AI-generated anchors have pioneered real-time voice and character animation synthesis in the news field, requiring only the input of the text content to be broadcast, allowing the computer to generate corresponding AI-generated anchor news videos, ensuring that the audio, expressions, and lip movements of the characters in the video remain naturally consistent, achieving an information delivery effect indistinguishable from that of real anchors.
2. AIGC + E-commerce: Promoting the Integration of Virtual and Real, Creating Immersive Experiences.
With the development and application of digital technology and the acceleration of consumer upgrades, the direction of e-commerce development is towards immersive shopping experiences. AIGC is accelerating the construction of 3D models of products, virtual anchors, and even virtual shopping venues, achieving immersive shopping experiences through the combination with new technologies like AR and VR, enabling multi-sensory interactions.
3. AIGC + Film and Television: Expanding Creative Space, Enhancing Work Quality.
With the rapid development of the film and television industry, process-related issues from pre-creation, mid-shooting to post-production have also become apparent, including a relative lack of high-quality scripts, high production costs, and the need to improve the quality of certain works, necessitating structural upgrades. The application of AIGC technology can stimulate creative ideas for film and television scripts, expand creative space for film and television characters and scenes, greatly enhancing the post-production quality of film and television products, and helping to maximize the cultural and economic value of film and television works.
4. AIGC + Entertainment: Expanding Radiation Boundaries, Gaining Development Momentum
In the digital economy era, entertainment not only brings products and services closer to consumers but also indirectly satisfies modern people’s desire for belonging, becoming increasingly important. With the help of AIGC technology, through interesting image or audio-video generation, creating virtual idols, and developing digital avatars for C-end users, the entertainment industry can quickly expand its radiation boundaries, gaining new development momentum in ways that are more easily accepted by consumers.
5. AIGC + Others: Promoting the Integration of Digital and Real, Accelerating Industrial Upgrades
In addition to the above industries, AIGC applications in education, finance, healthcare, and industry are also rapidly developing.
In the education sector, AIGC injects new vitality into educational materials. In finance, AIGC helps achieve cost reduction and efficiency increase. In healthcare, AIGC empowers the entire diagnostic and treatment process. In industry, AIGC enhances industrial efficiency and value.
Overall, AIGC is developing into a horizontal combination deeply integrated with various industries, and its related applications are accelerating penetration into all aspects of economic and social life.
Challenges Facing the Development of Artificial Intelligence Generated Content
As artificial intelligence technology enters the fast lane of development, AIGC plays an important role in various aspects of social production and life due to its rapid response capabilities, vivid knowledge output, and rich application scenarios. However, at the same time, the key technologies of AIGC, core competencies of enterprises, and relevant laws and regulations are not yet complete, and disputes surrounding fairness, responsibility, and safety are increasing. Key technologies are not fully mature, and there are still pain points and difficulties in large-scale promotion and implementation. Currently, AIGC technology is continuously upgrading, further unleashing content production capacity, but it still has limitations in critical AI technologies, which hinder the progress of industry development.
First, there are inherent flaws in artificial intelligence algorithms; second, AIGC content editing and creation technologies are not yet perfected; third, enterprise risk governance capabilities are still lacking.
In May 2022, the latest issued “Opinions on Promoting the Implementation of National Cultural Digitalization Strategy” called for the study and development of industrial policies to support cultural digitalization construction, emphasizing that localities should formulate specific implementation plans based on local conditions, and relevant departments should refine policy measures. In the future, the support intensity of policies from various regions and departments, their promotion and implementation, and dynamic adjustments will determine the degree of mutual construction between technology and society, which will play an important role in the development of AIGC technology in social contexts.
Future Outlook
From Real and Controllable to Diverse Combinations
From a technical perspective, the current AIGC-related algorithms have the ability to realistically replicate and create certain predetermined content, while related models have achieved good results in generating content for simple scenes. However, in the face of challenges in generating diverse and complex scene content, the existing AIGC algorithm capabilities still need further enhancement.
From Localized Centralization to Large-Scale Distribution
As a new model of deep integration between the digital economy and the real economy, AIGC creates and produces rich, innovative, high-quality, and interactive digital content through the application of a large number of new artificial intelligence technologies, presenting new challenges for current AI technology research. The research on large-scale distributed AIGC interaction algorithms is an urgent and popular topic today and is one of the future development directions of AIGC technology.
As AIGC’s core technology continues to develop, its foundational capabilities in content twins, content editing, and content creation will be significantly enhanced.
Currently, AIGC has been widely applied in various fields such as finance, media, entertainment, and e-commerce, and future application scenarios will further diversify.
Ecological Construction is Becoming Increasingly Perfect
With the continuous maturation of AIGC, an ecological system architecture centered on standards, technology R&D, content creation, industry applications, and asset services will become increasingly refined, whether empowering industrial upgrades with AIGC or autonomously releasing value through AIGC will develop healthily and orderly under this framework.
With the rapid development of cutting-edge technologies such as 5G, cloud computing, VR, and AR, and the research and innovation of a new generation of intelligent terminal devices, a complete AIGC ecological chain will be one of the most important driving forces in the future for unleashing data factor dividends, promoting traditional industry upgrades, advancing digital economy development, constructing a unified integration of digital and real, and creating a metaverse world.
This article is excerpted from the “White Paper on Artificial Intelligence Generated Content (AIGC)” by the China Academy of Information and Communications Technology and JD Exploration Research Institute.
