Analysis of AIGC Industry Prospects and Current Status: IDC Predicts Explosive Growth in AI Applications by 2024
-
In recent years, generative AI and large models have developed rapidly, even surpassing human capabilities in many fields, bringing great convenience as tools.
It should be noted that overuse of AI in research and other processes is not a good thing.
For example, researchers may misleadingly present AIGC-generated text as their own creation, or rely solely on AIGC and produce unreliable research results, which may even infringe on others' work.
Ministry of Science and Technology Issues Guidelines to Regulate AI Use, Prohibiting Direct Generation of Application Materials via AIGC
According to the Ministry of Science and Technology, the Supervision Department recently compiled and issued the "Guidelines for Responsible Research Conduct (2023)." The guidelines stipulate that generative AI (AIGC) must not be used to directly generate application materials, nor should generative AI be listed as a co-author of research outcomes. Additionally, the guidelines emphasize that researchers should integrate ethical requirements throughout the entire research process. The guidelines apply to research institutions, universities, medical and health institutions, enterprises, and their researchers, covering the main stages and processes of scientific and technological activities.
Regarding the publication of research achievements, the Guidelines emphasize that the release of breakthrough research results and major research progress must be approved by the respective research institution. Researchers are prohibited from disseminating to the public any research findings that have not undergone scientific verification or peer review. They must not republish already published papers or their data, images, etc., nor combine parts of multiple published papers to create "new achievements" for publication.
What are the main features of the Guidelines?
First, they reflect consensus. The Guidelines fully incorporate academic norms and behavioral standards long established and widely recognized by China's scientific community, while also drawing on beneficial international experiences and reflecting global practices.
Second, they emphasize comprehensiveness. The Guidelines apply to research institutions, universities, medical and health institutions, enterprises, and their researchers, with normative requirements covering the main stages and processes of scientific activities.
Third, they focus on practicality. The scientific ethics principles and academic research norms outlined in the Guidelines are meant to be universally followed by research institutions and researchers. The content and wording are designed to be concise, understandable, and implementable, ensuring strong operational feasibility.
Generative AI (AIGC - Artificial Intelligence Generated Content) is a significant marker of the transition from AI 1.0 to AI 2.0 era.
The accumulation and fusion of technologies such as GAN, CLIP, Transformer, Diffusion, pre-trained models, multimodal techniques, and generative algorithms have spurred the explosion of AIGC. Continuous iteration and innovation in algorithms, the qualitative leap in AIGC capabilities driven by pre-trained models, and the diversification of AIGC content through multimodal advancements have equipped AIGC with more universal and stronger foundational abilities.
From the progressive development of computational intelligence, perceptual intelligence, to cognitive intelligence, AIGC has opened the door to cognitive intelligence for human society. Through training on large-scale datasets, AI has acquired knowledge across multiple domains. With appropriate adjustments and refinements to the models, AI can now accomplish tasks in real-world scenarios.
AIGC Industry Prospects and Current State Analysis
Since early 2023, general artificial intelligence, represented by cognitive large models, has sparked a global frenzy. International companies like OpenAI, Microsoft, and Google have intensified their efforts, while domestically, a 'thousand-model battle' has erupted, with numerous high-tech enterprises racing to invest in R&D. According to the 'China Artificial Intelligence Large Model Map Research Report' released by the Ministry of Science and Technology, China and the U.S. dominate the global landscape of released large models, accounting for over 80% of the total.
Meanwhile, large models are also seen as a new engine for corporate transformation. McKinsey's report, 'The Economic Potential of Generative AI: The Next Productivity Frontier,' indicates that applying the 63 analyzed generative AI use cases across industries could add $2.6 trillion to $4.4 trillion annually to the global economy. This projection does not yet include all generative AI applications; if unstudied applications are factored in, the economic impact of generative AI could double.
Recently, at the 10th WAVE SUMMIT Deep Learning Developer Conference, Baidu's Chief Technology Officer and Director of the National Engineering Research Center for Deep Learning Technology and Applications, Wang Haifeng, announced that the user base of Wenxin Yiyan has surpassed 100 million. This milestone achievement signifies Baidu's significant progress in the field of artificial intelligence.
Wang Haifeng revealed that since its approval for public service on August 31, the number of user inquiries for Wenxin Yiyan has been steadily increasing, closely aligned with the performance improvements of the Wenxin large model. More and more users are trusting and utilizing Wenxin Yiyan, which fully demonstrates Baidu's technical strength and market competitiveness in the AI field.
Wang Haifeng disclosed at the conference that since its public launch on August 31 this year, the volume of user inquiries has been on a continuous rise, largely in sync with the enhancements in the Wenxin large model's effectiveness.
It was reported that Baidu did not start from scratch in developing generative AI products. The company possesses the necessary computing power, algorithms, and data for AI. Baidu has strategically deployed across four technological layers: chips, frameworks, models, and applications. Benefiting from these four layers, Baidu holds a comprehensive advantage in generative AI application technology, both in China and globally.
In November, Baidu released its unaudited financial report for the third quarter of 2023. The data showed that the company's revenue for the reporting period was 34.447 billion yuan, with Baidu's net profit (non-GAAP) reaching 7.3 billion yuan, a 23% year-on-year increase. Both revenue and net profit exceeded market expectations.
IDC Predicts: Explosive Growth in AI Applications Expected in 2024
DingTalk and IDC released the '2024 AIGC Application Layer Top 10 Trends White Paper.' With the development of AIGC technology, intelligent applications are expected to experience explosive growth. IDC predicts that by 2024, more than 500 million new applications will emerge worldwide. The white paper highlights that application layer innovation will be a definitive direction for AIGC industry development in 2024, AI Agents will become the mainstream form of large model deployment in business scenarios, and applications will transition from cloud-native to AI-native.
IDC also forecasts that by 2026, two-thirds of cloud applications will utilize AI, making it challenging for up to 80% of enterprises to find skilled AI professionals.