Chinese AI chatbot DeepSeek banned from government devices over security fearsBy political reportersMaani TruuandNicole HegartyTopic:Artificial IntelligenceTue 4 FebTuesday 4 FebruaryTue 4 Feb 2025 at 11:30amDeepSeek has been banned from all federal government devices, effective immediately.(Reuters: Florence Lo)In short:DeepSeek, a Chinese AI chatbot that made headlines late last month for rivalling ChatGPT, has been banned from all federal government devices.According to the government, the decision follows advice from national security and intelligence agencies that determined the platform posed "an unacceptable risk" to national security.What's next?Federal government workers will still be able to use the program on their personal devices.DeepSeek, the Chinese artificial intelligence chatbot that sparked an international frenzy last month, has been banned from federal government computers and mobile devices after it was found to pose "an unacceptable risk" to national security.The ChatGPT competitor was the number one downloaded app on Apple's app store early last week after it made headlines for wipingbillions of dollars from Wall Street.Under the new ban, all government bodies, except corporate organisations like Australia Post and the ABC, will be forced to remove all DeepSeek products from their devices effective immediately.They will also have to block access to DeepSeek products and report back to the government when they have completed it.Employees will still be able to use the program on their personal devices.DeepSeek isChina's first foray into the AI marketand the shock popularity of its latest model — released in January — quickly raised questions about data security and government censorship.According to the government, the decision follows advice from national security and intelligence agencies that determined the platform posed "an unacceptable risk to Australian government technology".Social media app TikTok was banned from government devices in 2023, due to similar concerns over security and the risk that it could be used for foreign interference. TikTok is also owned by a Chinese company, Byte Dance.Home Affairs Minister Tony Burke on Tuesday said the government was "country-agnostic" when it made decisions of this sort and instead "focused on the risk to the Australian government and our assets".DeepSeek is the first AI assistant developed by a Chinese-based company.(Reuters: Dado Ruvic)"AI is a technology full of potential and opportunity, but the government will not hesitate to act when our agencies identify a national security risk," he said.The announcement follows similar moves by Italy, Texas and parts of the US military.Government urges caution for all AustraliansDeepSeek was founded by 40-year-old entrepreneur Liang Wenfeng in 2023 and released its first large language model the same year.It has attracted global attention in part due to its claims that the model was far cheaper and took far less computing power to create compared to other AI products,turning the tech industry upside down.Science Minister Ed Husic was among the first Western leaders to warnthat there were "unanswered questions" about the platform's data and privacy managementlate last month.Dana Mckay, an associate professor at RMIT's School of Computing Technologies, said DeepSeek was required to feed the data it collects to the Chinese government."It does collect everything you do when you access DeepSeek," she said."Things like keystroke patterns and the way we type is actually as unique as a fingerprint so it can basically identify you."If the app is installed on your computer, she said, it is able to ask for root or administrator access, "which would mean it could access pretty much everything on your computer".The government's special envoy for cyber security, Andrew Charlton, said the software raised "serious concerns around data security and potential vulnerabilities in government systems".Mr Charlton said while the ban only applies to government devices, the public should take note."The government's advice is that people should be vigilant when using DeepSeek," he said."We're advising people not to click on links which could lead to malware, not to upload any information that they don't want to be public and not to rely on the accuracy of any information provided by generative AI."DeepSeek has been contacted for comment.