Alibaba shares plummeted to a 16-month low amid the AI scandal involving Anthropic

Alibaba Group shares fell to a 16-month low during trading in Hong Kong on June 25. This came after AI startup Anthropic accused the Chinese internet giant of a campaign to gain “illegal” access to its Claude model using thousands of fake accounts, Bloomberg reported. Anthropic sent an official letter regarding the incident to several U.S. senators and White House officials, asking them to take action.

Alibaba's stock price plummeted 4.9%, bringing its year-to-date decline in market capitalization to 33%. Against this backdrop, shares of other Chinese developers of large language models also fell: Xiaomi and Baidu each lost more than 3%.

In a letter reviewed by Bloomberg, Anthropic claims that the attack was carried out by operators linked to the Qwen AI lab, which is owned by Alibaba. The campaign targeted Claude’s most valuable capabilities—specifically, software development and the logical reasoning of AI agents, the document states. Between April and June, 28.8 million requests were made to Claude through nearly 25,000 fake accounts as part of this scheme, Anthropic writes. The AI startup emphasized that this is Alibaba’s largest attempt to date to exploit the technological infrastructure of leading U.S. labs.

The scale of the problem has raised serious concerns in Silicon Valley: Anthropic, OpenAI, and Google have already joined forces to share information about such incidents, Bloomberg adds. Previously, Anthropic and OpenAI had accused other Chinese startups, including DeepSeek and Minimax, of using distillation—a method in which developers train their neural network using the responses and results of another.

Anthropic’s allegations are heightening investor concerns: the ability of Chinese tech giants to compete in the global AI race is now in question, Bloomberg explains. At the same time, local players have so far managed to offer the market quite competitive products at affordable prices.

Doubts about Alibaba’s success in the AI market continue to weigh on its stock price as the company struggles with weakening consumer demand in China, according to Bloomberg. The company is effectively redirecting most of the profits from its Chinese e-commerce business toward attracting and expanding the user base for its AI model, the agency reported earlier.

The situation is further exacerbated by a global rotation of capital: investors are actively pulling out of the Chinese internet sector and shifting their funds into stocks of semiconductor and equipment manufacturers in South Korea and Taiwan.

At the same time, regulatory pressure from Washington is mounting: in June, U.S. authorities ordered Anthropic to block foreign users from accessing its most advanced platforms. This came after a vulnerability was discovered in the latest Fable 5 model that allowed users to bypass built-in restrictions.

“Chinese AI models face an increased risk of being banned by the U.S.,” noted Robert Li, an analyst at Bloomberg Intelligence. “The White House’s ban on foreign access to Anthropic’s flagship models, Fable 5 and Mythos 5, signals a shift toward a more aggressive regulatory policy on AI in the U.S.”

This article was AI-translated and verified by a human editor