Trump Adviser Accuses Chinese AI Firm of Stealing Anthropic Tech
Photo: Mark König
A senior technology adviser to Donald Trump has alleged that China’s Moonshot AI misappropriated technology from the American startup Anthropic.
A high-ranking technology adviser to Donald Trump has ignited a geopolitical firestorm by accusing a prominent Chinese artificial intelligence firm of intellectual property theft. The allegations center on Moonshot AI, one of China’s most highly valued technology startups, which is accused of improperly obtaining and utilizing proprietary code developed by the American AI research company, Anthropic.
During a recent briefing, the adviser suggested that the sophisticated architecture powering Moonshot’s large language models bears striking similarities to the foundational research produced by Anthropic, a leading US-based developer known for its focus on AI safety and alignment. These claims, if proven, represent a significant escalation in the ongoing global race for artificial intelligence dominance, further straining the already fragile technological relationship between Washington and Beijing.
The accusation comes at a time when the United States is increasingly concerned about the security of its homegrown technology. American policymakers have long warned that Chinese firms may be leveraging unauthorized access to advanced software and hardware designs to accelerate their own development cycles, bypassing years of costly research and development. Moonshot AI, founded in 2023, has rapidly become a central figure in China’s domestic AI landscape, securing substantial investment from major players like Alibaba and Tencent. The startup has focused on creating high-performance chatbots and AI infrastructure capable of competing with the global market leaders.
Anthropic, for its part, has not publicly confirmed the specific details of the alleged breach but has previously emphasized the rigorous security protocols it maintains to protect its intellectual property. The company is backed by tech giants like Amazon and Google, and its technology is considered a cornerstone of the Western AI ecosystem. Protecting such innovation has become a national security priority for the US government, which has implemented strict export controls on advanced AI chips and software tools to prevent them from reaching Chinese firms.
Industry analysts note that the software architecture of large language models can be notoriously difficult to protect. Because these models are trained on massive datasets and utilize complex mathematical frameworks, distinguishing between independent innovation and unauthorized replication can be a complex technical challenge. However, the Trump adviser’s intervention suggests that intelligence agencies may have evidence pointing toward a more deliberate transfer of technology. The claims follow broader patterns of scrutiny surrounding Chinese AI development, with Western officials frequently expressing skepticism regarding the rapid progress of domestic Chinese models.
As the allegations gain traction, they are expected to fuel calls for tighter regulations and increased oversight of international research collaborations. The tech sector is watching closely, as any move to blacklist Chinese AI firms or restrict further investment could disrupt global supply chains and significantly alter the market dynamics for venture capital funding. For investors, the situation adds a new layer of risk, as companies operating in the crosshairs of US-China technological competition face mounting regulatory uncertainty.
The implications of this dispute extend beyond corporate rivalry. It underscores the broader ideological struggle over the future of artificial intelligence, a technology viewed by both Washington and Beijing as the key to economic and military superiority in the 21st century. As governments continue to treat AI development as a matter of national defense, the boundary between commercial software development and international espionage remains increasingly blurred. Whether these specific allegations lead to formal legal action or new executive orders remains to be seen, but they clearly signal that the 'AI cold war' is entering a more aggressive phase.
This article was generated based on trending topic: “China's Moonshot AI stole from Anthropic, Trump tech adviser says - BBC”