DeepSeek Exposed for Leaking Sensitive Data
Cybersecurity firm Wiz (USA) has reported discovering a cache of sensitive data belonging to DeepSeek that was unintentionally exposed on the Internet.
In a blog post, Wiz stated that DeepSeek left over a million lines of unsecured data accessible online, including digital software keys and chat logs that appeared to record user queries sent to the company’s free AI assistant.

DeepSeek software interface
According to Reuters, Ami Luttwak, Wiz’s Chief Technology Officer, stated that DeepSeek swiftly took action to secure the data after his company alerted them.
“They took it down in less than an hour,” Luttwak said, but he emphasized that the sensitive data was so easily accessible that he was unsure if his team was the first or only group to discover it.
DeepSeek has yet to comment on Wiz’s blog post.
China’s groundbreaking AI model, DeepSeek, has been making waves in the tech industry in recent days, surpassing renowned competitors such as ChatGPT, Gemini, and Claude AI in several benchmark tests. The company is highly regarded for its significantly lower research costs compared to major industry players, as well as its short development timeline.
DeepSeek was founded by Liang Wenfeng in May 2023 and is headquartered in Hangzhou, Zhejiang. It is owned by the investment firm High-Flyer. Funded by High-Flyer, DeepSeek has no plans for external fundraising, focusing instead on developing foundational AI technology. According to ChinaTalk, unlike many AI companies in China, DeepSeek claims its mission is to “decode the mystery of artificial general intelligence (AGI) through curiosity.” The company’s laboratory is currently researching innovations in architecture and algorithms that have the potential to revolutionize the field of artificial intelligence.