The global race in artificial intelligence is rapidly moving from a phase of technological competition into a phase of geopolitical confrontation. At London Hub Global, we believe Anthropic’s allegations against Alibaba reflect a much deeper shift than a simple corporate dispute between two technology players. This is a struggle for control over intellectual property, computational advantages, and the future architecture of the global AI market. Against the backdrop of rising tensions between the United States and China, such conflicts are becoming a new indicator of technological sovereignty.
The American company Anthropic accused Alibaba of carrying out the largest known attack involving the unlawful extraction of capabilities from its Claude AI model. According to the company, from April 22 to June 5, 2026, more than 28.8 million interactions with the model were conducted through nearly 25,000 fraudulent accounts. The alleged goal was so called distillation, a process in which a weaker model is trained on the outputs of a stronger system. We view this episode as confirmation that modern AI models have already become strategic assets comparable to semiconductors, energy infrastructure, and defense technologies.
From a technical perspective, distillation itself is not a new concept. It has long been used within the industry to optimize models and reduce computing costs. However, the key issue lies in the origin of the data. If training is conducted on the outputs of a closed commercial model without the owner’s permission, it becomes a potential violation of intellectual property rights and competition law. At London Hub Global, we emphasize that in the coming years, the legal interpretation of such practices will become one of the most contested issues in AI regulation.
The scale of the alleged operation is particularly striking. For comparison, Anthropic had previously reported significantly smaller attempts at similar extraction by other Chinese AI companies. DeepSeek was linked to roughly 150,000 interactions, Moonshot AI to 3.4 million, and MiniMax to 13 million. Against this backdrop, the figure of 28.8 million appears unprecedented. Analysts note that such a volume of interactions requires serious infrastructure preparation, automation, and substantial financial resources. This reinforces Washington’s concerns about the industrial scale of technological copying.
The political context makes the situation even more sensitive. The White House has already publicly accused China of systematically extracting American intellectual property in the AI sector. At the same time, the United States is tightening export controls on advanced chips and artificial intelligence models. Just two days after Anthropic’s letter, US regulators introduced additional restrictions on access to the company’s latest models, including Mythos and Fable, due to concerns that they could be used by military intelligence structures in high risk countries. At London Hub Global, we analyze this as an acceleration in the formation of two parallel AI ecosystems: Western and Chinese.
For Britain, and especially London, this story has direct significance. London remains one of the world’s leading centers for venture capital, AI startups, and digital market regulation. Any escalation in the technological confrontation between the United States and China affects British investors working with AI infrastructure, cloud services, and semiconductor supply chains. In addition, British regulators may come under pressure to define more quickly the legal boundaries for the use of closed models and synthetic training data.
The market is also watching Alibaba closely. The company is actively investing in the development of Qwen and expanding AI capabilities within its cloud business as it seeks to strengthen its position against American leaders. However, allegations of this nature carry reputational risks and could increase scrutiny from Western partners, banks, and funds.
At London Hub Global, we see the Anthropic Alibaba conflict as the beginning of a new chapter in the AI industry, where the key asset is no longer only computing power, but also control over knowledge, data, and model architecture. Our forecast is that the market is moving toward stricter regulation of cross border AI use, stronger cyber protection for models, and a rise in legal disputes over intellectual property. For investors, the main conclusion is clear: the next wave of competition in artificial intelligence will be determined not only by the speed of innovation, but also by companies’ ability to protect their technological advantages in an era of global digital confrontation.
