In just a few years, ChatGPT has emerged and DeepSeek has shaken the world. These big models stir up the "spring water" in the field of artificial intelligence - the United States has vowed to build a global capital of artificial intelligence, and the EU has vowed not to be a supporting role in the artificial intelligence competition. How to view the current global competition landscape in artificial intelligence? How can China seize the "bull's nose" of artificial intelligence and explore its unique development path? The reporter interviewed several members of the Chinese People's Political Consultative Conference who are currently attending the National People's Congress and the Chinese People's Political Consultative Conference. "At present, the overall development of China's artificial intelligence has become the first echelon in the world." Qiao Hong, an academician of the CAS Member and director of the multimodal AI system laboratory, said that with the support of system, policy, talent and market advantages, with the further integration of innovation chain and industrial chain, China's artificial intelligence field will usher in greater development opportunities. This year's government work report clearly proposes to continue promoting the "Artificial Intelligence+" action, better combining digital technology with manufacturing and market advantages, and supporting the widespread application of large models. Qi Xiangdong, Chairman of Qianxin Group, believes that the rise of China's open-source big models such as DeepSeek fully demonstrates the innovation vitality and strength of China's AI, and promotes the global AI industry towards a more open and win-win future. DeepSeek has achieved world-class performance with low computing power through algorithm innovation and deep parallel optimization Zhang Yunquan, a researcher at the Institute of Computing Technology of the Chinese Academy of Sciences, said that this does not mean that the scale rule has failed. In view of the current hot discussion about the choice of technical paths, such as computing power or computing efficiency, Wang Liang, a researcher at the Institute of Automation, Chinese Academy of Sciences, believes that ChatGPT and DeepSeek have their own advantages and disadvantages, and they are not antagonistic, but complementary. Currently, developed countries are accelerating the layout of a new generation of intelligent computing centers. How to break through the siege of computing power has become an urgent issue for the development of artificial intelligence in China. Computing power, especially high-end computing infrastructure, remains a key strategic resource for global competition. "Zhang Yunquan suggested building a national level computing power cluster, introducing management measures for the construction of high-end intelligent computing centers, concentrating resources to support leading enterprises in large models, and creating a world leading open source sovereign level basic universal large model. Many committee members also gave their own thoughts and suggestions on how to achieve China's AI self-reliance. Zhao Xiaoguang, a researcher at the Institute of Automation of the Chinese Academy of Sciences, told the reporter: "We should pay more attention to original innovation and speed up the seizure of the world's scientific and technological commanding heights." "The key to China's AI victory in the future depends on the combination of the national system and mass innovation." Qi Xiangdong believed that we should further strengthen top-level design and forward-looking layout, implement targeted support policies, stimulate the innovation and creativity of the whole society, and then help AI key technologies achieve greater breakthroughs. Qiao Hong suggests that we should increase our attention and investment in basic research, continuously strengthen the foundation of key core technologies, enhance collaborative innovation among government, industry, academia, research, and application, further improve the level of scientific research achievement transformation, and build an open and integrated ecological innovation model. China has rich application scenarios for artificial intelligence. Zhang Yunquan believes that we should encourage the release of more high-value scenarios, relax application access, carry out industry pilot projects, and promote application demonstration and promotion. While deepening the practical application of artificial intelligence, it is also necessary to continuously study and sort out the problems encountered in the application Zhao Xiaoguang said that it is necessary to timely formulate governance norms, expand the scope of technical service scenarios and markets, and gain more experience and consensus in serving the world. In terms of enhancing the discourse power of rule making, Wang Liang proposed to adhere to open source and support domestic enterprises and research institutions to participate in the international open source community, promote Chinese technology standards to become international standards, enhance China's discourse power in global artificial intelligence governance, especially in the fields of data privacy, algorithm transparency, and artificial intelligence ethics, actively participate and propose Chinese solutions, and strive to promote the establishment of fair and just global governance rules. (New Society)