Image Source: Generated by Wujie AI

Recently, hundreds of AI applications in China have been removed from the App Store, attracting widespread attention from the industry. With more and more large-scale model enterprises and AI applications emerging, the healthy and sustainable development of the domestic large-scale model industry has become a topic of increasing concern and contemplation.

Recently, Liu Cong, Vice President and Dean of the Research Institute of iFLYTEK, pointed out in an interview with Sina Technology that although China's artificial intelligence industry has made breakthrough progress, it still lacks a unified and controllable generation framework and self-detection standards for generated content. With the emergence of ethical and security issues brought about by the landing of large-scale models, it is urgent for China to strengthen the standardization of large-scale models. By formulating unified national standards, the quality and performance of large-scale models can be ensured to meet the expected level, safeguarding user data security and privacy protection to support the healthy and sustainable development of the large-scale model industry.

In Liu Cong's view, the healthy development of the large-scale model industry in China urgently requires standardization.

Recently, the list of units represented by the joint leaders of the large-scale model topic group of the National Artificial Intelligence Standardization Overall Group was announced, with Liu Cong, Vice President and Dean of the Research Institute of iFLYTEK, representing the company as the joint leader of the topic group. According to Liu Cong, iFLYTEK has attached great importance to standardization work since its establishment, establishing a standardization management committee at the group level with the chairman as the director, and the company's leadership team and department heads serving as committee members to unify the planning and management of various standardization work within the group.

"Liu Cong stated, "As early as 2003, iFLYTEK served as the leading unit of the original Chinese Speech Interaction Standard Working Group of the former Ministry of Information Industry; from 2013 to the present, Chairman Liu Qingfeng has served as the deputy director of the National Information Technology Standardization Committee User Interface Sub-Technical Committee, and the company serves as the leading unit of the Speech Interaction Standard Working Group; from 2018 to the present, Senior Vice President Hu Guoping has served as the deputy leader of the National Artificial Intelligence Standardization Overall Group; and from 2022 to the present, I have served as the deputy leader of the National Information Technology Standardization Committee Artificial Intelligence Sub-Technical Committee Knowledge Graph Working Group."

He introduced that as of now, iFLYTEK has led and participated in the formulation of more than 70 intelligent speech and artificial intelligence standards at home and abroad, including 2 international standards, 29 national standards, 8 industry standards, and 9 group standards that have been published and implemented, covering various fields such as speech recognition, speech synthesis, intelligent home, intelligent customer service, mobile terminals, car terminals, smart TVs, and machine translation.

Now, with the re-election as the joint leader of the large-scale model topic group of the National Artificial Intelligence Standardization Overall Group, iFLYTEK's participation in the formulation of national artificial intelligence standards has added a new identity.

In Liu Cong's view, standards are the technical support for economic activities and social development, and an important aspect of the country's fundamental system. Standardization plays a fundamental and leading role in the country's governance system and modernization construction governance capacity. Currently, China's artificial intelligence industry has made breakthrough progress, but still lacks a unified and controllable generation framework and self-detection standards for generated content. With the emergence of ethical and security issues brought about by the landing of large-scale models, "it is urgent to strengthen the standardization of large-scale models, to ensure the quality and performance of large-scale models reach the expected level, safeguard user data security and privacy protection, and support the healthy and sustainable development of the large-scale model industry."

Amid the "chaotic dance" of large-scale models, the competition is a long-term battle.

According to incomplete statistics from Sina Technology, as of now, over 100 large-scale model products have been launched in China, with as many as 26 companies positioning themselves as either general large-scale models or "general + industry" enterprises. In addition, there are 7 "basic + industry" large-scale model companies, and 67 large-scale models in other industries such as finance, medical, biology, transportation, and data analysis.

With more and more large-scale model enterprises and AI applications emerging, how can the domestic large-scale model industry develop in a standardized manner? Liu Cong sees that in this competition sparked by ChatGPT, domestic large-scale model enterprises striving to catch up with the advanced level abroad are destined to encounter various difficulties and obstacles in the process of "charging straight ahead," such as challenges in chips, high-quality data, system iteration, and regulation. "Charging straight ahead" will also be a "long-term battle."

Liu Cong believes that industry must take the lead on this road, driving the overall development of the ecosystem after innovative applications. Of course, maintaining momentum is also crucial for benchmarking and continuously narrowing the gap. After achieving phased results, industrial development and scientific research innovation must advance in tandem, continuously iterating and promoting each other's paradigms to continuously inject long-term driving force into the development of independent large-scale models in China.

Confident in the development of national standards for large-scale models, iFLYTEK will do three things in the future.

Since its establishment, iFLYTEK has been engaged in core technology research such as speech and language, natural language understanding, machine learning reasoning, and autonomous learning, and has achieved outstanding results in various international evaluations such as machine translation, natural language understanding, image recognition, image understanding, knowledge graph, knowledge discovery, and machine reasoning.

"Based on our company's strong technical foundation and strength, we have the ability and confidence to do a good job in the research and development of national standards for large-scale models. We will fully leverage our own technological and R&D advantages, actively communicate and cooperate with all parties, and jointly promote the formulation and improvement of national standards for large-scale models. At the same time, we will continue to explore and innovate to provide technical support for the high-quality development of the artificial intelligence industry," said Liu Cong.

He introduced that under the guidance of the National Artificial Intelligence Standardization Overall Group and the large-scale model topic group, iFLYTEK will do the following work for participating in the formulation of national standards for large-scale models:

First, provide technical support and share experience. iFLYTEK has deep technical accumulation and rich application experience in the fields of intelligent speech, image, NLP, etc., and will provide technical support and share experience for the formulation of national standards, actively promote the landing application and promotion of national large-scale model standards in various industries such as finance, medical, education, and transportation, in order to better understand and solve practical problems.

Second, promote the establishment and improvement of a scientific and systematic evaluation system. iFLYTEK has long been committed to the research and innovation of artificial intelligence technology, and will actively strengthen the research and innovation of large-scale model evaluation technology and methods, improve the scientific and objective nature of evaluation, establish a professional evaluation team and a sound evaluation system, provide scientific evaluation results and suggestions for the formulation of national standards, and promote the continuous innovation and development of large-scale model technology.

Third, strengthen cooperation and communication. iFLYTEK will strengthen cooperation and communication with other joint leader units, relevant enterprises, research institutes, universities, etc., to jointly promote the construction and monitoring of the large-scale model ecosystem, collect and analyze data and feedback information of large-scale models in different scenarios, and ensure the rationality and operability of the formulation of national standards.

免责声明：本文章仅代表作者个人观点，不代表本平台的立场和观点。本文章仅供信息分享，不构成对任何人的任何投资建议。用户与作者之间的任何争议，与本平台无关。如网页中刊载的文章或图片涉及侵权，请提供相关的权利证明和身份证明发送邮件到support@aicoin.com，本平台相关工作人员将会进行核查。