
Department of Homeland Security cites national security risks, data transfer fears, and potential vulnerabilities in latest crackdown on Chinese tech.
The Trump administration banned Chinese AI startup DeepSeek from U.S. civilian government agencies, citing national security concerns, data transfer risks, and cybersecurity vulnerabilities amid intensifying technological tensions between Washington and Beijing.
AI-generated summary
The Trump administration issued an order barring civilian U.S. government agencies from using DeepSeek technology due to security concerns, following earlier restrictions by Congress and other bodies.
The Trump administration banned Chinese AI startup DeepSeek from civilian U.S. government agencies, citing security concerns.
The Department of Homeland Security on Monday made public the order requiring federal civilian agencies to remove DeepSeek-related technology from their IT systems and barring them from procuring DeepSeek products and services. The ban applies to federal civilian agency IT systems, including third-party products and services developed using the company's technology.
The move comes amid intensifying geopolitical tensions between Washington and Beijing, with artificial intelligence emerging as a key battleground in the two countries' race for technological dominance.
Asked about the U.S. move by Reuters, China's foreign ministry told the news agency that it would safeguard the rights and interests of its companies.
Mao Ning, a ministry spokesperson, said that "China consistently opposes the over-generalization of national security as well as the politicization of science, technology and trade issues," the agency reported.
The order, originally issued on July 24, applies to federal civilian agencies, including the Treasury, State and Justice departments, but it excludes the Department of Defense.
Since DeepSeek burst onto the scene at the start of 2025, several U.S. government agencies have imposed various restrictions.
Congress restricted the use of DeepSeek in intelligence community systems in December 2025. Earlier that year in March, U.S. Commerce Department bureaus told staff DeepSeek was banned on government devices, Reuters reported at the time. The Defense Information Systems Agency moved to block access to DeepSeek models in January 2025, according to reports from Bloomberg.
DeepSeek's AI models "present multiple threat vectors," the latest DHS order states, citing concerns over potential access to sensitive data by Chinese authorities and potential cybersecurity vulnerabilities.
The agency also warned that using DeepSeek's AI models could result in sensitive U.S. government data being transferred to China, potentially providing intelligence on American citizens and government programs.
The models may also contain malware or other security vulnerabilities, the agency said, citing open-source research that identified multiple critical vulnerabilities in DeepSeek's iOS app.
Executive agencies can request exceptions to the order, including delays in compliance, by submitting a written justification to the DHS within 60 days of the order's issuance.
AI outlook — possibilities, not facts
Executive agencies will submit exception requests to the DHS within 60 days.
Likely · Within months

Apple announced a product launch event on October 28 in Los Angeles, with invitations sent to media and analysts. The event is expected to focus on high-end Pro devices, including potential new Macs and a touchscreen notebook, marking a shift in launch cadence under CEO John Ternus. Apple updated Mac Mini and Mac Studio in August and MacBook Pro in spring, but iMac remains unchanged since October 2024. Macs now represent 10% of revenue and are central to Apple's AI strategy, bolstered by a recent partnership with Meta to bring an open-weight model to Mac Studios. Apple shares are up 26% year-to-date, outperforming the S&P 500.

AI developer Anthropic has announced it will commit 5% of its shares, valued at around $48bn, to a public good fund ahead of its anticipated stock market listing.

Character.AI has filed for Chapter 11 bankruptcy as part of a structured deal allowing Disney to hire most of its staff and license its technology while leaving the startup responsible for ongoing lawsuits and state regulatory investigations.

OnlyFans creators express frustration over Meta allowing its Muse AI agent to engage in explicit adult conversations while human creators face strict account restrictions and censorship on Instagram.

Facing growing public pushback against massive data center campuses, the data center industry is increasingly targeting K-12 schools with children's books, flag-making kits, field trips, and educational programs to build community goodwill and corporate storytelling.

The World Bank is urging developing nations to adopt "small AI" models that operate on mobile phones with low connectivity to accelerate economic growth and address local needs, as discussed at the IMF and World Bank meetings in Bangkok.