Tuesday, April 28, 2026

Kim Jong Un’s Economic Vision: Why Coal and Cement Are Key to North Korea’s 5-Year Plan

Kim Jong Un highlights coal and cement industries during elections, emphasizing economic development in North Korea's new five-year plan.

Revolutionary InP Nanosemiconductors: How KAIST Boosted Light Efficiency to 18.1%!

KAIST's breakthrough in semiconductor tech enhances indium phosphide nanocrystals' light efficiency from under 1% to 18.1% through atomic-level control.

THE CHINA AI SHOCK: US Tech Giants Are NOW Officially LOSING The AI Race To Budget Open-Source Models

Chinese AI firms like DeepSeek and Alibaba are outperforming U.S. models, capturing significant global market share with open-source technologies.

U.S. Warns Against Chinese AI DeepSeek’s Distillation Tactics and Theft

PoliticsU.S. Warns Against Chinese AI DeepSeek's Distillation Tactics and Theft

The U.S. State Department has dispatched a diplomatic cable to its embassies worldwide, sounding the alarm on Chinese companies’ efforts to pilfer American artificial intelligence (AI) technology, including those of AI startup DeepSeek, Reuters reported on April 24.

According to the cable obtained by Reuters, the State Department clarified that the message aims to highlight the risks associated with using AI models that have distilled proprietary American AI models, and to lay the groundwork for future U.S. government actions and external communications.

Distillation is a technique that trains smaller models using the output of large AI models, allowing for the development of powerful AI tools at a fraction of the cost.

The cable specifically named DeepSeek, along with other Chinese AI firms such as Moonshot AI and MiniMax.

In the message sent on April 18, State Department officials were instructed to convey concerns about hostile entities extracting and distilling U.S. AI models to their foreign counterparts. The cable also mentioned that a separate demarche and message had been delivered to China.

The State Department emphasized that models developed through unauthorized and covert distillation of American AI models enable foreign entities to launch products with comparable performance on certain benchmarks at significantly lower costs. They argued that such actions deliberately compromise security protocols and undermine the ideological neutrality and truth-seeking capabilities of AI models.

Related to this issue, OpenAI warned in February that DeepSeek was targeting ChatGPT developers and major U.S. AI companies to replicate and train its models.

On April 23, Michael Kratsios, Director of the White House Office of Science and Technology Policy, announced on X (formerly Twitter) that evidence had been uncovered of foreign companies, including those from China, conducting large-scale distillation campaigns to steal American AI technology. He pledged to take action to safeguard U.S. innovation.

However, the Chinese embassy in Washington dismissed these allegations as baseless, asserting that China places great importance on intellectual property rights protection, Reuters reported.

DeepSeek, which made waves last year with its low-cost AI models, recently unveiled a preview version of a new model optimized for Huawei’s chip technology, further showcasing China’s growing capabilities in AI self-sufficiency.

Check Out Our Content

Check Out Other Tags:

Most Popular Articles