In-depth Market Research on AIGC Industry and Analysis of AIGC Current Status
-
The Concept of AIGC
Generative AI (AIGC) is a pivotal milestone in the evolution from AI 1.0 to AI 2.0. The integration of technologies like GAN, CLIP, Transformer, Diffusion, pre-trained models, multimodal techniques, and generative algorithms has driven the rapid expansion of AIGC. Continuous algorithmic innovation, the qualitative leap in AIGC capabilities enabled by pre-trained models, and the diversification of AIGC content through multimodal approaches have endowed AIGC with more versatile and robust foundational capabilities.
From the progression of computational intelligence to perceptual intelligence and then to cognitive intelligence, AIGC has opened the door to cognitive intelligence for human society. By training on large-scale datasets, AI can acquire knowledge across multiple domains, requiring only minor adjustments to perform tasks in real-world scenarios. The core idea of AIGC technology is to use AI algorithms to generate creative and high-quality content. Through model training and learning from vast datasets, AIGC can produce content based on input conditions or guidance. For example, given keywords, descriptions, or samples, AIGC can generate matching articles, images, audio, and more.
Centered around large models, AIGC offers services such as modeling tools, security services, content detection, and foundational platforms. The upstream segment of the AIGC industry chain primarily provides AI technologies and infrastructure, including data suppliers, data analysis and labeling, creator ecosystems, and related algorithms. The midstream focuses on vertical sectors like text, images, and videos, offering data development and management tools, including content design, operational efficiency enhancements, and data organization. The downstream includes content end markets, content services and distribution platforms, digital assets, smart devices, and AIGC content detection.
Analysis and Research on the Current State of AIGC Industry Development
AIGC represents a milestone for both human society and artificial intelligence. In the short term, AIGC transforms fundamental productivity tools; in the medium term, it reshapes social production relations; and in the long term, it drives qualitative breakthroughs in overall societal productivity. In this transformation of productivity tools, production relations, and productivity, the value of data as a production factor is greatly amplified. AIGC elevates data to the status of a core resource of the era, accelerating the digital transformation of society to some extent.
The enhancement of human resource services by AIGC is currently one of the fastest-growing areas in corporate management systems. It significantly improves the efficiency of human resource management while also transforming traditional HR management models, such as the three-pillar framework. For instance, in recruitment, AIGC's core value lies in resume recommendations, empowering stages like screening, interview filtering, and assessments. By analyzing candidate resumes, generating matching reports, and intelligently recommending suitable candidates based on company needs, AIGC greatly enhances screening accuracy and efficiency, reducing the workload of HR departments.
With the continuous iteration of deep learning models, AIGC has achieved groundbreaking progress. Particularly in 2022, algorithms experienced explosive growth, and breakthroughs in foundational technologies made AIGC commercialization feasible. Key developments include the AI art field: in June 2014, Generative Adversarial Networks (GAN) were introduced; in February 2021, OpenAI released the CLIP multimodal pre-trained model; and in 2022, Diffusion Models gradually replaced GAN.
The AI industry chain consists of three main layers: the foundational layer, the technical layer, and the application layer. The foundational layer focuses on building support platforms, including sensors, AI chips, data services, and computing platforms. The technical layer emphasizes core technology R&D, covering algorithm models, foundational frameworks, and general-purpose technologies. The application layer concentrates on industrial applications, including industry-specific solutions, hardware, and software products. Research indicates that China's AIGC industry chain comprises five components: foundational large models, industry/scenario-specific mid-models, business/domain-specific small models, AI infrastructure, and AIGC support services, forming a robust industrial ecosystem.
Currently, AIGC technology and applications in China are concentrated in areas such as marketing, office work, customer service, human resources, and basic operations, with the value and empowerment of this technology already preliminarily validated. 33% of enterprises prioritize AIGC for marketing, 31.9% for online customer service, 27.1% for digital office scenarios, and 23.3% for IT and security enhancements.
AIGC is an AI technology built on multimodal capabilities, enabling a single model to understand language, images, videos, and audio simultaneously and perform tasks beyond the reach of single-modal models, such as adding captions to videos or generating images based on semantic context. In China, AIGC currently appears mostly as single-model applications, categorized into text generation, image generation, video generation, and audio generation, with text generation serving as the foundation for other content types.
The commercialization of AIGC abroad starts with foundational large models, with typical applications like ChatGPT and Midjourney being incubated based on the invocation of these foundational models. In contrast, the domestic market, characterized by its highly diverse business scenarios and fragmented supply-side services, has seen AIGC commercialization begin with smaller, domain-specific models. Foundational large models are still in a phase of rapid iteration and upgrades, while also beginning to focus on specific business scenarios. The market for industry or scenario-specific models is relatively lagging, but in the context of China's unique market environment, this will become an area where both foundational large models and domain-specific small models actively cross boundaries.
Currently, the primary issue is the lack of comprehensive laws and regulations related to AIGC. To fully leverage AIGC technology, it is essential to improve the relevant legal frameworks. From the current applications of AIGC technology, there is a lack of well-defined safety standards, and the legal and social responsibilities of AIGC technology services, content dissemination, and technical applications are not clearly outlined. Additionally, the absence of robust legislation and hierarchical regulatory measures for AIGC technology makes it difficult to ensure its security.
In the use of AIGC technology, there is no clear delineation between public and proprietary data, leading to issues with the compliance, security, and ownership of data used in training foundational large models. For example, the leakage of proprietary data may compromise user data security, while data elements struggle to realize their full value. Technical confidentiality is a paramount concern for AIGC. For instance, during interactions with AIGC, proprietary resources of enterprises may be leaked. Insufficient technical confidentiality could severely impact the owners of information resources.
As AIGC technology continues to evolve and its applications expand, it is recommended to gradually improve the legal and regulatory framework to ensure its healthy development. This includes establishing a legal access system, conducting research on laws and regulations for AIGC model market entry, and clarifying the legal and social responsibilities of AIGC technology services, content dissemination, and technical applications. Furthermore, encouraging multi-stakeholder participation in legislative research, implementing hierarchical regulatory measures, and fostering public-private partnerships in industry governance are essential steps forward.