The smartphone apps DeepSeek’s site is seen on a laptop screen in Beijing, Tuesday, Jan. 28, 2025. Andy Wong/AP
Andy Wong/AP
Washington and Europe are becoming more watchful of DeepSeek.
Last weekend, the Taiwanese artificial intelligence company surprised the world by appearing to be less expensive than the popular robot ChatGPT. However, authorities and privacy advocates are presently posing fresh concerns about the security of user data.
As the state investigates the company’s collection and storage of data, Italian regulators have banned the app from Apple and Google app stores it.
Authorities in and are looking into whether the AI robot poses a privacy risk.
Members of Congress and their employees are also being warned by the House’s Chief Administrative Officer not to use the software in the United States. In an effort to “outcompete” China in the development of AI and” safeguard Americans ‘ data,” two U.S. representatives the Trump administration to tighten existing restrictions on the sale of semiconductor chips to China.
What information is DeepSeek obtaining?
According to DeepSeek’s protection plan, the company collects a mine of consumer data, including chat and search query history, the device a user is on, key patterns, IP addresses, internet connection and activity from other apps.
Another AI companies, like OpenAI’s ChatGPT, Anthropic’s Claude, or Perplexity, produce a similar volume of data from customers. But do social media apps like Facebook, Instagram and X. At days, these kinds of data collection techniques have led to questions from officials.
Angela Zhang, a law professor at the University of Southern California with a focus on Chinese legislation, said that “data security concerns are usually a critical issue when using AI bots, and this is not only a problem with DeepSeek.” She cited the EU’s stringent data protection laws, which are applied even to AI businesses with US bases like OpenAI, which have faced substantial investigation and studies in the past. For instance, these require people to settle in to any data set.
Where is customer files kept by DeepSeek?
DeepSeek sends all the data it collects on Americans to machines in China, according to the bank’s .
And for security experts, that is where the difficulty lies.
Much like Washington’s fears about TikTok, which prompted Congress to boycott the game in the U. S., the problem is that a China-based business will eventually be answerable to the state, probably exposing Americans ‘ sensitive data to an adversarial state.
( The Trump administration is currently implementing the TikTok ban. ) The president issued an to extend the policy’s effective meeting until April.
According to Samm Sacks, a study professor who studies Chinese security at Yale, the chatbot may in fact pose a threat to American national security.
” That data, in overall, can be used to gain insights into a community, or customer behaviors that could be used to create more effective phishing attacks, or other malicious adjustment campaigns”, Sacks said.
Has the Chinese government used DeepSeek to access Americans’ data?
There are no reports of Chinese officials using DeepSeek to obtain personal information about Americans.
The concerns about China are hypothetical, with the only chance that Beijing might use American data to raise questions, much like the TikTok debate.
According to Yale’s Sacks, the potential data risk presented by DeepSeek can be compared to two other important factors.
First, the Chinese government already has an unfathomable amount of data on Americans.
In December, Chinese hackers the U. S. Treasury Department’s computer systems. After infiltrating U.S. telecom companies, a different group of Chinese hackers last year on Americans ‘ texts and calls. Not to mention that a sizable amount of data on Americans is frequently purchased and sold by a sizable network of digital .
China is already very capable of obtaining Americans ‘ data, even without a potential backdoor like DeepSeek, according to Sacks.
And secondly, DeepSeek is open source, meaning the chatbot’s software code can be viewed by anyone. On top of the underlying code, developers can also create their own apps and services.
Because businesses firmly guard the software code as confidential intellectual property, the most well-known AI chatbots are not open source. Sacks contends that DeepSeek’s transparency of how data is accessed and processed acts as a system check.
However, American privacy experts remain unsure of what happens to the data once it is stored on Chinese servers. In China, this would require the company to comply with national security regulations, which would require the government to provide it with information.
Requests for comment to DeepSeek, and its owner, Chinese hedge fund High-Flyer, were not returned.
Does DeepSeek censor answers?
Researchers and journalists have discovered instances in which DeepSeek has censored sensitive political subjects while simultaneously Chinese propaganda.
While competing chatbots have no trouble explaining the 1989 Tiananmen Square massacre, DeepSeek told NPR:” Sorry, that’s beyond my current scope. Let’s talk about something else”.
Similar to how DeepSeek responded to a question about Taiwan’s sovereignty, which China claims is a part of its territory, before the answer vanished and was replaced with the statement that the question is beyond DeepSeek’s purview.
Experts say the chatbot’s limits are a reminder that the internet is government-controlled in China and that tech companies like DeepSeek are subject to interference, so the chatbot’s answers should be greeted with a heavy dose of skepticism.
Is there any safe way to use DeepSeek?
Privacy experts, like Sacks, claim that there are ways to use DeepSeek more safely.
They advise against signing up for the app through Gmail or Apple accounts to ensure that DeepSeek’s data is still accessible.
It’s also best to avoid asking questions that reveal identifying or highly personal information when using the chatbot because all data entered into DeepSeek is used to train the AI model, which is ultimately stored in China.
People can also use a virtual private network, a VPN, to hide their location.
She said avoiding DeepSeek completely is the safest bet if you work for the government or if your personal information is potentially sensitive.
However, in the end, users are ultimately responsible for protecting digital data in the U.S. because there are no federal data privacy laws written on the books, unlike the European Union, granting citizens insight and control over how their digital data is being accessed and used by private companies. Instead of creating national regulations, Sacks said,” We’re kind of playing Whack-A-Mole with this rotating cast of apps from China”.
She continued,” We can ban TikTok, we can raise the red flag of data security concerns for DeepSeek, but we don’t have a more comprehensive way to make sure that businesses aren’t gathering more data than they should,” adding:” We don’t have a more comprehensive way to ensure that companies aren’t collecting more data than they should.