US artificial intelligence (AI) company Anthropic PBC said on Monday that it had uncovered campaigns by three Chinese AI firms to illicitly extract capabilities from its Claude chatbot, in what it described as industrial-scale intellectual property theft.
Anthropic said DeepSeek (深度求索), Moonshot AI Technology Co (月之暗面) and MiniMax Group Inc (稀宇科技) used a technique known as “distillation” — using outputs from a more powerful AI system to rapidly boost the performance of a less capable one.
“These campaigns are growing in intensity and sophistication,” the company said in a statement. “The window to act is narrow, and the threat extends beyond any single company or region.”
Photo: Bloomberg
Distillation is a common practice within AI development, often used by companies to create cheaper, smaller versions of their own models.
The practice grabbed headlines a year ago when the release of a low-cost generative AI model from DeepSeek performed at a similar level to ChatGPT and other top US chatbots, upending assumptions of US dominance in the sensitive sector.
Anthropic said the Chinese companies achieved their ends through approximately 16 million exchanges with its Claude model and 24,000 fake accounts. These allowed the three labs to siphon off capabilities they had not independently developed, at a fraction of the cost — and in so doing circumvented export controls on powerful US technology intended to preserve US dominance in AI.
The San Francisco-based company argued the practice posed national security risks, saying models built through illicit distillation are unlikely to retain safety guardrails designed to prevent misuse — such as restrictions on helping develop bioweapons or enabling cyberattacks.
Anthropic’s archrival, OpenAI, creator of ChatGPT, made similar accusations to US lawmakers earlier this month, saying Chinese companies were using the technique amid “ongoing efforts to free-ride on the capabilities developed by OpenAI and other US frontier labs.”
Other US officials, including White House AI adviser David Sacks, have also expressed concerns that DeepSeek used this method.
Anthropic said MiniMax ran the largest operation, generating more than 13 million exchanges.
Each campaign concentrated heavily on coding, agentic reasoning and tool use — areas in which Claude is considered a leader.
To circumvent Anthropic’s ban on commercial access from China, the labs allegedly routed traffic through proxy services that managed the vast networks of fraudulent accounts.
Anthropic called for coordinated industry and government responses to address what it said no single company could tackle alone.
European Central Bank (ECB) President Christine Lagarde is expected to step down from her role before her eight-year term ends in October next year, the Financial Times reported. Lagarde wants to leave before the French presidential election in April next year, which would allow French President Emmanuel Macron and German Chancellor Friedrich Merz to find her replacement together, the report said, citing an unidentified person familiar with her thoughts on the matter. It is not clear yet when she might exit, the report said. “President Lagarde is totally focused on her mission and has not taken any decision regarding the end of
French President Emmanuel Macron told a global artificial intelligence (AI) summit in India yesterday he was determined to ensure safe oversight of the fast-evolving technology. The EU has led the way for global regulation with its Artificial Intelligence Act, which was adopted in 2024 and is coming into force in phases. “We are determined to continue to shape the rules of the game... with our allies such as India,” Macron said in New Delhi. “Europe is not blindly focused on regulation — Europe is a space for innovation and investment, but it is a safe space.” The AI Impact Summit is the fourth
CONFUSION: Taiwan, Japan and other big exporters are cautiously monitoring the situation, while analysts said more Trump responses ate likely after his loss in court US trading partners in Asia started weighing fresh uncertainties yesterday after President Donald Trump vowed to impose a new tariff on imports, hours after the Supreme Court struck down many of the sweeping levies he used to launch a global trade war. The court’s ruling invalidated a number of tariffs that the Trump administration had imposed on Asian export powerhouses from China and South Korea to Japan and Taiwan, the world’s largest chip maker and a key player in tech supply chains. Within hours, Trump said he would impose a new 10 percent duty on US imports from all countries starting on
STRATEGIC ALLIANCE: The initiative is aimed at protecting semiconductor supply chain resilience to reduce dependence on China-dominated manufacturing hubs India yesterday joined a US-led initiative to strengthen technology cooperation among strategic allies in a move that underscores the nations’ warming ties after a brief strain over New Delhi’s unabated purchase of discounted Russian oil. The decision aligns India closely with Washington’s efforts to build secure supply chains for semiconductors, advanced manufacturing and critical technologies at a time when geopolitical competition with China is intensifying. It also signals a reset in relations following friction over energy trade and tariffs. Nations that have joined the Pax Silica framework include Japan, South Korea, the UK and Israel. “Pax Silica will be a group of nations
RSS