The United States may become the second country after Australia to ban Chinas DeepSeek artificial intelligence on government devices.
U.S. Representatives Josh Gottheimer and Darin LaHood introduced a bipartisan bill proposing the ban.
In their letter to 47 U.S. governors and the mayor of Washington, the congressmen warned that DeepSeek could pose security risks to sensitive government data and cybersecurity and Americans privacy, NBC News reported on March 3.
China denies the allegations. However, concerns highlighted by the U.S. lawmakers and state officials are not without merit, experts say.
The Chinese government has reportedly also used AI models for mass surveillance, including the collection of biometric data and social media listening operations that report to Chinas security services and the military, as well as for information attacks on U.S. and Chinese dissidents abroad.
At least three leading Chinese surveillance and security companies — TopSec, QAX and NetEase — announced the integration of DeepSeek to enhance their services.
TopSec and QAX provide services directly to the Chinese government, and NetEase made it clear that DeepSeek will improve their cyber censorship and surveillance capabilities.
Over the past decade, the Chinese government has been investing heavily in AI-driven biometric data capturing, face recognition and surveillance technologies such as smart cities, the Skynet project, and the Sharpe Eyes program, which can monitor all aspects of an individuals public life, Wenhao Ma of VOAs China Division reported.
In January, Canadian cybersecurity firm Feroot Security uncovered a code imbedded in DeepSeeks login processes that shares user information with Chinese state-owned communication company China Mobile, AP reported.
The Associated Press described the code as a heavily obfuscated computer script that when deciphered shows connections to computer infrastructure owned by China Mobile.
The U.S. banned China Mobile in 2019 following intelligence reports that it serves as the Chinese militarys spy arm.
China-based actors have been using ChatGPT along with DeepSeek models to generate phishing email and disinformation attacks on the U.S. on behalf of unspecified clients in China, OpenAI said in its February report.
OpenAI identified and blocked a cluster of China-originated accounts involved in malicious activities, such as Qianyue Overseas Public Opinion AI Assistant, reportedly designed to ingest and analyze posts and comments related to Chinese politics and human rights from platforms such as X, Facebook, YouTube, Instagram, Telegram and Reddit.
The purpose of the operation was reportedly to feed the resulting insights to the Chinese authorities such as Chinese embassies abroad, and to intelligence agents monitoring protests in countries including the United States, Germany and the United Kingdom, OpenAI said.
A set of ChatGPT accounts that OpenAI banned in February had been involved in Chinese influence operations focused on generating short comments in English and long-form Spanish-language articles critical of the United States published in local and national media outlets across Latin America and Spain.
One of the Chinese companies planting the articles in the Spanish-language outlets was Jilin Yousen Culture Communication Co., a subsidiary of the government-tied Beijing United Publishing House.
VOA reviewed nine of the Chinese AI-generated articles published in Spanish-language media between October and November 2024 as identified by OpenAI.
Two — in Mexicos El Universal and Perus El Popular — criticized the United States use of sanctions targeting foreign governments and individuals.
The El Universal op-ed described the U.S. sanctions on Irans oil industry for Tehrans backing of terrorist groups Hamas and Hezbollah as exposing the U.S. impotence in dealing with globa
