Earlier this March when Chinese delegates descended on Beijing for the Two Sessions, an annual meeting of the People’s Republic’s two rubber-stamp legislative bodies, there was much excitement about artificial intelligence (AI). DeepSeek, China’s answer to U.S. AI companies such as OpenAI was the talk of town. Chinese officials appear to have sensed an opportunity to advance a wider goal that had already been on the minds of PRC leadership: the adoption of AI across various sections of the Chinese economy, in particular industrial applications—or AI+, as it has come to be referred to in Chinese political discourse.
Xinhua, China’s state-run news agency reporting from the Two Sessions, celebrated AI+ as key to China’s technological rise and its ability to prevail in a “new round of scientific and technological revolutions.” Other state media reports followed suit in the weeks following the Two Sessions and described other specific applications of artificial intelligence such as “AI+ SME,” i.e., how AI might help boost small to medium enterprises.
This year’s government work report, a document presented every year at the Two Sessions by China’s premier, sketched out a renewed push for the “AI+ Initiative.” “Under the AI+ initiative, we will work to effectively combine digital technologies with China’s manufacturing and market strengths. We will support the extensive application of large-scale AI models and vigorously develop new-generation intelligent terminals and smart manufacturing equipment, including intelligent connected new-energy vehicles, AI-enabled phones and computers, and intelligent robots,” Premier Li Qiang declared.The general idea is that all the technologies likely to define technological leadership in the coming years should get an AI boost through broad applications across these industries.
The AI+ Initiative is not new. It was launched during last year’s Two Sessions and included in the 2024 government work report. At that time, DeepSeek had yet to become a global sensation. Instead, the success of ChatGPT which made a splash in 2022 continued to be a thorn in the sides of Chinese officials fixated on winning the technological race with the United States. Now, the DeepSeek moment offered an opportunity to give the initiative another push.
Pushing for Industrial AI
The announcement of such plans has a certain tradition, and those proclaiming them are always in search of momentum: Since China’s first-ever 2017 “AI Action Plan,” the country has been on the lookout for moments to push initiatives forward that risked falling behind under the immense pressure of their big goals. Also, China’s President Xi Jinping is known for seeing technological moments as particularly pivotal to the fate of the People’s Republic of China.
The DeepSeek moment is one such tech moment, which, China’s leaders hope, will galvanize the AI+ Initiative that already triggered various local-level and company-specific efforts. The Province of Henan published a two-year “AI+ Plan” in early 2024. State-owned enterprises such as China Mobile also joined the fray last year. State media has also been reporting on “AI+ healthcare” and “AI+ education” scenarios; the latter seeks to collect and synthesize learning data using AI.
On top of this, DeepSeek how has unlocked excitement on the part of the Chinese Communist Party (CCP) leadership about the prospects of AI even fixing one of China’s immediate economic problems. Another recent “AI+” combination appears aimed at spurring Chinese consumption. The State Council issued a plan to boost consumption including a measure called “AI+ consumption.” In a Xinhua interview, Hong Tao, Director of the Beijing Technology and Business University Institute of Business Economics, expressed his belief that the effort could help Chinese consumers “to want to consume, to be able to consume, to know how to consume, and to enjoy consuming.” In this way, AI+ is being stretchedto also drive economic activity through domestic consumption.
The Quest for Technological Leadership
One issue that will continue to generate significant debate is the question of what constitutes technological leadership. When on February 5, 2025, The Global Times, one of the CCP’s most forceful international propaganda mouthpieces, celebrated DeepSeek’s success. “DeepSeek achieved a breakthrough under low computational power, which once again validates the paradox of ‘sanctions-driving innovation.’” Yet, such statements do not settle the debate about AI leadership.
Among those attending the Two Sessions were key figures from China’s corporate world, and it is their responsibility to give interviews on the sidelines of the event to talk up the latest political trends. In a Two Sessions interview with Qiushi, a theoretical journal of the CCP, Liu Qingfeng, founder of iFlyTek and a deputy to the 14th National People’s Congress, had the following to say: “Objectively speaking, domestically made large language models are still some ways away from high-level self-reliance and complete independent controllability.” Liu identifies the problem the following way: “At this point in time, the vast majority of domestic large-model training is highly dependent on imported computing power. This brings a risk akin to building a house on someone else’s foundation.” The message was clear: China continued to depend on the import of high-end chips from the US, irrespective of the DeepSeek win.
Talking about the ability to “control” technology “independently” is a common reference to the ongoing tech denial strategy the United States is pursuing. “High-level self-reliance,” meanwhile, is a term born out of an even deeper desire to not only be able to manufacture technological products without American inputs, but for new technologies to emerge from China-based research activities. In the same interview, Liu proceeded to highlight the need for greater investment in AI research and cited his own company’s efforts as an example. This means that there is a difference between broad adoptability of AI and cutting-edge model development. Both could count as leadership in AI, depending on your definition.
Disentangling these various signals is no easy task. The challenge for Europe, however, is clear: European leaders need to be ready to respond by committing—over the long term—to fostering a vibrant AI ecosystem in Europe. There is much to be optimistic about here, given the plethora of European initiatives the European Commission and member states have announced over the past several months. The key will be to not let up too soon. Chinese state planners certainly won’t.
Michael Laha is a senior research fellow with the German Council on Foreign Relations’ (DGAP) Center for Geopolitics, Geoeconomic, and Technology.
