U.S. Accuses China of 'Industrial-Scale' AI Theft, Beijing Fires Back

Introduction: U.S.-China AI Rivalry Enters New Phase of Tension

The rivalry between the United States and China over artificial intelligence technology is entering an entirely new phase of tension. The U.S. recently publicly accused China of systematically stealing core American AI technologies on an "industrial scale," spanning model architectures, training data, and key algorithms. China swiftly responded, characterizing the accusations as "slander" and emphasizing that its AI achievements are built entirely on independent innovation.

Meanwhile, with a highly anticipated Trump-Xi summit on the horizon and Washington preparing a series of sweeping sanctions, further uncertainty is being added to an already strained bilateral relationship.

Core Event: 'Industrial-Scale Theft' Accusations Trigger Diplomatic Storm

According to multiple international media reports, senior U.S. government officials have explicitly used the term "industrial-scale" to describe China's alleged theft of American AI technology. Washington claims that China obtains AI research and technology from U.S. companies and research institutions through multiple channels, including but not limited to cyber intrusions, talent recruitment programs, technology transfers through academic collaborations, and investment infiltration of American AI startups.

The U.S. has outlined several specific concerns: persistent cyberattacks targeting major American tech companies, aimed directly at core training data and model weights of large language models; the channeling of cutting-edge research to domestic institutions through Chinese researchers based in the United States; and exploitation of gray areas within the open-source ecosystem to reverse-engineer and modify restricted technologies.

In response, a Chinese Ministry of Foreign Affairs spokesperson took a firm stance, calling the U.S. claims pure "slander" and "fabrications." China stressed that its achievements in artificial intelligence are the result of the hard work of millions of scientific researchers, and that any attempt to discredit China's technological progress is irresponsible. Beijing also pointed out that the United States' longstanding technology blockade and chip bans against China are the real actions undermining the global order of scientific and technological cooperation.

Sanctions Loom: Bargaining Chips Before the Summit

Adding further complexity to the situation, sources familiar with the matter have revealed that the U.S. government is considering a new round of sweeping sanctions against China. These potential measures could include an expanded AI-related entity list, financial restrictions on Chinese AI companies, and further tightening of advanced chip export controls to China.

Analysts believe that Washington's decision to release these signals on the eve of the Trump-Xi summit is both a pressure tactic and a bid to gain more leverage at the negotiating table. However, this approach carries significant risks — if sanctions are too severe, they could substantially diminish the summit's outcomes or even cause further deterioration in bilateral relations.

Notably, the scope of the sanctions under discussion is unprecedented. Beyond traditional areas such as chips and semiconductor equipment, the U.S. is also considering extending restrictions to AI cloud computing services, Model-as-a-Service (MaaS) platforms, and AI-related venture capital activities. If these measures are ultimately implemented, they will have far-reaching consequences for the global AI supply chain.

Deep Analysis: The Essence of the Battle for Technological Sovereignty

From a broader perspective, the "theft" accusations reflect a fundamental competition between the U.S. and China over AI technological sovereignty.

First, in terms of technological capabilities, the speed of China's progress in AI has genuinely alarmed Washington. Chinese AI companies such as DeepSeek have achieved remarkable breakthroughs in large language models, with some performance benchmarks approaching or even matching world-leading levels. This rapid advancement appears "implausible" to the U.S. side, providing important context for the accusations.

Second, the open-source versus closed-source debate is another focal point of contention. The American tech community itself is divided on the question of open-sourcing AI models. On one hand, companies like Meta actively promote open-source strategies; on the other, there are concerns about adversaries leveraging open-source models. Chinese companies' extensive use of open-source models for fine-tuning and improvement occupies what some in the U.S. view as a gray area between legitimate use and "theft."

Third, talent mobility remains a highly sensitive issue. In recent years, multiple "espionage" cases involving Chinese-American scientists have ultimately been proven to be wrongful prosecutions, damaging academic freedom and creating a severe chilling effect in the research community. However, the latest U.S. accusations suggest that the underlying policy direction has not fundamentally shifted.

Additionally, this round of competition has extended well beyond the bilateral level. The United States is actively rallying allies to build an "AI democracy alliance," seeking to isolate China's AI ecosystem on a global scale. The cooperation of Japan, the Netherlands, and other nations on chip equipment export controls is a concrete manifestation of this strategy.

Industry Impact: Global AI Ecosystem Faces Risk of Fragmentation

If the U.S.-China AI confrontation continues to escalate, the global artificial intelligence industry will face an unprecedented risk of fragmentation. A U.S.-led AI ecosystem and a China-centered AI ecosystem could accelerate into formation, profoundly affecting global tech companies' strategic planning, supply chain management, and market choices.

For China's AI industry, the short term may bring greater pressure on technology access, but in the long run, this will further drive the advancement of independent innovation capabilities. From "chokehold technologies" to "autonomous and controllable" solutions, China's domestic substitution efforts in chips, computing infrastructure, and AI frameworks are expected to accelerate.

For the global AI research community, this trend toward geopoliticization is deeply concerning. Restrictions on cross-border academic collaboration, obstacles to paper sharing, and barriers to talent mobility will all slow the pace at which artificial intelligence benefits humanity as a whole.

Outlook: At the Crossroads of Confrontation and Cooperation

The upcoming Trump-Xi summit will serve as a critical barometer for the direction of U.S.-China relations. Whether the two sides can find a balance between competition and cooperation in the AI domain will not only shape the technological futures of both nations but will also profoundly influence the global landscape of artificial intelligence governance.

On the optimistic side, topics such as AI safety, ethical standards, and global governance frameworks still offer room for collaboration. On the pessimistic side, if sweeping sanctions are implemented, a U.S.-China AI "decoupling" could shift from trend to reality.

Regardless of the summit's outcome, one thing is already abundantly clear: artificial intelligence has become a central battleground in great power competition, and the trajectory of this contest will largely define the global technological order of the 21st century.