離開
進入《TAIPEI TIMES》 Government agencies banned from using DeepSeek, digital ministry says
The DeepSeek logo is displayed on a smartphone screen with a Chinese flag in the background on Wednesday. Photo: Reuters
By Jonathan Chin / Staff writer, with CNA and AFP
The Ministry of Digital Affairs （MODA） on Friday said that public sector workers would be banned from using DeepSeek, a recently released Chinese artificial intelligence （AI） model, due to concerns over data leaks.
DeepSeek launched its R1 chatbot last month, claiming it matches the capacity of AI pacesetters in the US for a fraction of the investment.
Government agencies and critical infrastructure should not use DeepSeek, because it “endangers national information security,” MODA said in a news release on Friday.
“DeepSeek AI service is a Chinese product,” the ministry said. “Its operation involves cross-border transmission, and information leakage and other information security concerns.”
MODA said the ban applies to employees of central and local government agencies, public schools, state-owned enterprises and other semi-official organizations, as well as those working on critical infrastructure projects and in government-endowed foundations, but did not elaborate on specifics or enforcement measures.
MODA cited the Principles on Restricting the Use of Products That Endanger National Cyber Security, a 2019 Executive Yuan regulation which prohibits government agencies from using computer and information communication technology products and services that might pose a national security risk.
Separately, Democratic Progressive Party （DPP） Legislator Wang Ting-yu （王定宇） said DeepSeek is only the latest in a long list of Chinese applications, Web sites, technology and hardware banned from government use.
Taiwan is among the most exposed nations to the military and cybersecurity risks from China and should be as vigorous in confronting those threats as the US Navy, which has banned its personnel from using DeepSeek, he said.
Chinese telecoms, devices and hardware are banned from being connected to Taiwanese infrastructure due to concerns that they might contain spyware and backdoors, he said, adding that the US, European nations and India have discovered malicious Chinese codes.
Government offices and workers are prohibited from using DeepSeek, a rule that aims to protect national security and cybersecurity, he said, adding that people and private enterprises should also be mindful of the issue.
Government agencies should hasten to put into force a ban on high-risk tech products from China, including DeepSeek, TikTok and other software, hardware and device components, DPP Legislator Lin Chun-hsien （林俊憲） said.
Taiwan should move faster than other nations in responding to DeepSeek, as the nation is more threatened by Beijing than others, he added.
Taiwan’s restriction came as data watchdogs in South Korea and Ireland said they would ask DeepSeek to clarify how it manages users’ personal information.
Other countries, including Australia, Italy, Japan, the UK and the US, have also raised questions about the Chinese AI start-up’s data practices.
US secretary of commerce-designate Howard Lutnick accused DeepSeek of stealing US technology including Nvidia chips and codes before calling for punitive tariffs.
“I think they [China] only care about themselves and seek to harm us, and so we need to protect ourselves,” he was cited by US newspaper The Hill as saying during his confirmation hearing on Wednesday.
US representatives John Moolenaar and Raja Krishnamoorthi, respectively chairman and ranking member of the US House of Representatives Select Committee on the Strategic Competition between the US and the Chinese Communist Party, urged US President Donald Trump to tighten export restrictions on AI chips to China to counter DeepSeek’s rise.
Separately, the US Navy has forbidden service members from using the Chinese-developed AI model in any form.
British Secretary of State for Science, Innovation and Technology Peter Kyle has told Politico that London is conducting a national security risk assessment on the DeepSeek chatbot.
Meanwhile, Italian data protection agency Garante said it imposed restrictions on DeepSeek’s use of Italian nationals’ personal information and issued letters to the Chinese company’s offices to demand an explanation on how its algorithm obtains data.
DeepSeek has been removed from Google and Apple’s app stores in Italy.
Kyodo news agency has reported that Japanese Prime Minister Shigeru Ishiba told the Japanese Diet that he is mulling a basic law on generative AI.
新聞來源：TAIPEI TIMES
