Market Data 
In 2017, China unveiled an ambitious strategy for the domestic development of artificial intelligence (AI), setting a bold goal of establishing itself as the world’s leading AI innovation hub by 2030. With the aim of establishing itself as a global leader in AI by advancing AI theory, technology, and applications across sectors such as industry, governance, and defense, China is accelerating the construction of a global powerhouse in science and technology, in line with the directives of the CCP Central Committee and the State Council. Following the release of its AI strategy, China is already ranked number one globally in 2020, in terms of the number of research papers on AI and the number of AI related patents. China’s swift advance in AI was also helped by the size of its domestic market and its skilled workforce.
China’s AI industry has grown rapidly, with an annual growth rate exceeding 10% between 2018 and 2022. The China Academy of Information and Communications Technology (CAICT) reported that the core AI industry’s value reached RMB 508 billion (US$75 billion) in 2022, marking an 18% year-on-year increase. Cities like Beijing, Shanghai, and Shenzhen are leading this transformation, leveraging vast markets and diverse sectors.
In 2023, Shenzhen dominated the upstream AI market, leading in data, computation resources, and hardware infrastructure, with Beijing and Shanghai close behind. The city has prioritized AI integration in consumer technology, focusing on smartphones, smart computers, and XR devices, while also advancing digital humans and AI agents. Its strength in hardware and fast-paced commercialization cements its position as a major force in the country’s AI sector.
Beijing excels in core algorithm development and foundational R&D, supported by a strong talent pool and industrial ecosystem, making it a top destination for tech investors. Its AI strategy emphasizes autonomous innovation and large-scale applications in robotics, education, healthcare, culture, and transportation. Through benchmark projects, the city aims to advance large-model technology and enhance industry service quality.
Shanghai, a key center for AI applications in manufacturing, healthcare, and finance, provides a stable environment for scaling AI within a regulated market. The city has built a strong AI cluster, fostering coordinated innovation across industries and institutions. It is also expanding intelligent computing infrastructure, with major developments in areas like Songjiang and Lingang.
China’s long-term AI strategy, supported by adaptive regulations and infrastructure, ensures sustainable growth by integrating AI into industries like healthcare, manufacturing, and energy. With over 4,300 AI companies, China demonstrates how nations can align strategy, innovation, and policy to maximize AI’s potential. State-backed financial support, exemplified by initiatives from the Bank of China to provide at least 1 trillion Yuan in funding to companies in the Artificial Intelligence (AI) sector over the next five years to further fuel AI expansion. China’s AI ecosystem thrives on infrastructure, data, talent, and innovation. Investments in 5G networks and energy-efficient data centers provide essential computing power for large-scale AI models. A rapidly expanding data pool, bolstered by initiatives like the National Data Administration, ensures seamless data accessibility and interoperability. Meanwhile, AI-focused education and research programs are equipping the workforce with critical skills, reinforcing China’s commitment to global AI leadership.
Breakthroughs in AI technology like Large Language Models (LLM), particularly after ChatGPT-3’s release, have driven major Chinese tech firms—including Baidu, Alibaba, Tencent, and ByteDance—to invest heavily in large language models (LLMs). Baidu launched Ernie Bot just four months after ChatGPT-3, and China has since developed at least 240 LLMs through collaborations between tech giants, startups like Minimax and Zhipu AI, and academic institutions such as Peking and Tsinghua universities. Even non-tech companies like Xiaomi and Meituan have entered the AI space, integrating AI into their products and services.
DeepSeek trained its AI model for under $6 million using 2,000 chips—far less than Meta, which spent significantly more with over 16,000 chips. Large Language Models (LLMs) enable AI-driven applications across industries, advancing natural language processing and understanding. Unlike traditional domain-specific models, LLMs serve as scalable, general-purpose AI solutions, reducing infrastructure costs while driving innovation. Within days of its launch, DeepSeek AI’s mobile chatbot topped Apple’s App Store, surpassing OpenAI’s ChatGPT in downloads. This surge impacted financial markets, triggering a sell-off in major U.S. tech stocks, including Nvidia, Microsoft, and Meta.
Backed by former hedge fund High-Flyer, DeepSeek has built a reputation for innovation. Its V2 model, launched in May 2024, gained attention for its high performance and competitive pricing. The latest R1 model introduces key advancements, including large-scale reinforcement learning to enhance reasoning, a rule-based reward system that outperforms traditional neural models, and distillation techniques that compress AI knowledge into models as small as 1.5 billion parameters. Additionally, its Emergent Behavior Network showcases how complex reasoning can develop naturally through reinforcement learning.
With a quality score of 80, DeepSeek matches top models from OpenAI, Google, and Anthropic while offering significantly lower costs—just $0.9 per million tokens. Its affordability makes it an attractive choice for businesses and researchers seeking high-performance AI without excessive expenses.
Chinese investors are aggressively entering AI-related stocks, seeing DeepSeek’s rise as a pivotal moment in the Sino-U.S. tech rivalry. Stock surges in Chinese chipmakers, software firms, and data center operators have been fueled by patriotic sentiment amid renewed U.S. trade tariffs under President Trump. The Hang Seng AI Index jumped over 5% this week, while chipmaker and IT indices surged over 11%. Mainland investors returning from the Lunar New Year holiday have driven tech sector gains, bolstering AI, semiconductor, big data, and robotics stocks. Analysts view DeepSeek’s breakthrough as a response to U.S. attempts to limit China’s AI development. Goldman Sachs forecastsv that Chinese AI advancements could significantly impact stock markets, potentially boosting corporate earnings by 2% and increasing valuations by 20%, narrowing the gap with U.S. firms. With AI poised to transform industries, investors like Yi Xiangjun of Shenzhen Black Stone Asset Management (world’s largest alternative asset manager) are betting on China’s AI and tech sector, seeing it as an epoch-making revolution.
Concerns about AI’s impact on national security have been emphasized at the highest levels of the Chinese government. In May, President Xi Jinping called for a “new pattern of development with a new security architecture” to address AI’s “complicated and challenging circumstances”.
The rapid advancement of AI in China has been accompanied by stringent government oversight and censorship measures. Among the 40 policy and regulatory initiatives under the New Generation AI Development Plan launched in 2023, the “Interim Measures for the Management of Generative AI Services” require developers to register their algorithms with the Chinese government and undergo a “security assessment” if their services possess “social mobilization ability” capable of influencing public opinion thus regulating online conversations. The plan mandates adherence to “core socialist values” and specifies various restrictions on AI applications. While some measures aim to safeguard citizens—such as bans on promoting terrorism and disseminating “obscene pornography”—others serve to maintain government control over AI technology. Companies and users are prohibited from employing generative AI to “subvert state power,” “damage the country’s image,” or “undermine national unity”.
China asserts that AI played a crucial role in controlling the COVID-19 pandemic, aiding in tracking infections, forecasting trends, and facilitating economic recovery. A March 2021 study by the Mercator Institute for China Studies (MERICS) highlights how China’s social credit system, which integrates AI technology, was rapidly adapted for these purposes. The study traces the system’s origins to the 1990s as part of financial credit rating efforts, noting its continued significance in finance today. AI is also widely used in Chinese business sectors such as e-commerce and industrial automation. However, MERICS warns that technological advancements have significantly enhanced online surveillance, with near-ubiquitous camera coverage and extensive AI-driven public security monitoring.
Non-governmental organizations, including Human Rights Watch, have raised concerns over China’s AI-driven surveillance, particularly in Xinjiang, where such technologies have been linked to severe human rights abuses. Reports also indicate that China exports its surveillance tools to countries such as Ecuador. The 2017 AI plan explicitly calls for expanding AI applications in public safety, including the construction of intelligence-based monitoring, early warning, and control systems.
China’s rapid advancement in AI has positioned it as a formidable competitor to the United States, which has long been considered the global leader in the field. This progress was a key factor behind the U.S. Congress establishing the National Security Commission on Artificial Intelligence (NSCAI) in 2018. The NSCAI’s final report, published in March 2021, raised concerns about China’s AI capabilities, warning that “for the first time since WWII, America’s technological predominance—the backbone of its economic and military power—is under threat”. Recent reports have also highlighted allegations against Chinese companies, often linked to the government, of engaging in data theft. These accusations involve hacking foreign companies to steal sensitive intellectual property and trade secrets, particularly in the manufacturing, energy, and pharmaceutical sectors. Many of these activities are attributed to state-sponsored cyber operations known as “Advanced Persistent Threats” (APTs)”.
In a notable move, Texas has become the first U.S. state to officially ban the Chinese-owned AI app DeepSeek, citing concerns that the Chinese Communist Party (CCP) could access sensitive user data. Reports indicate that DeepSeek stores user data on Chinese servers, raising fears that the Chinese government could demand access under its data-sharing laws. Consequently, several U.S. government agencies, including the Navy and NASA, have already banned DeepSeek from their military and government devices. Concerns over DeepSeek’s data practices have triggered regulatory actions worldwide. Italy’s Data Protection Authority imposed a nationwide ban, removing the app from Apple and Google stores while investigating its compliance with European data laws. Taiwan’s Ministry of Digital Affairs banned the app on government devices and advised critical-sector businesses to do the same. In Australia, Home Affairs Minister Tony Burke declared DeepSeek a national security threat on February 4, announcing a nationwide ban. South Korea’s Ministry of Industry temporarily restricted employee access, urging caution with generative AI, while state-owned Korea Hydro & Nuclear Power had already blocked it earlier in the month. Additionally, South Korea’s privacy watchdog is set to investigate the app’s data management practices.
While concerns over data security persist, China’s AI industry has emerged as a global powerhouse, driven by groundbreaking innovations like DeepSeek’s cost-efficient models. With a vast talent pool and AI deeply integrated across industries—from healthcare to finance and manufacturing—China’s rapid advancements challenge Western dominance and reshape the technological landscape. Its ability to scale AI applications efficiently, backed by cutting-edge infrastructure and strategic investments, underscores its growing influence in the global AI race.
Rishab Rathi is serving as a Research Associate at the Center for Policy Research and Governance (CPRG). The views and opinions expressed in this article are those of the author.
Read the full article here
