The Chinese AI startup DeepSeek unintentionally exposed its complete stash of sensitive data due to an online exposure discovered by the cybersecurity firm Wiz.
Wiz shared in a blog post that DeepSeek servers exposed more than one million lines of unprotected data, application keys, and user-initiated prompts sent to the company’s free AI assistant.
Activities to protect the sensitive data began following an alert to DeepSeekfrom Wiz’s technology officer, Ami Luttwak. He told Reuters that the data disappeared from servers within “less than an hour.” He added, “This was so simple to find, we believe we’re not the only ones who found it.”
DeepSeek has not commented on this matter yet. After launching its virtual assistant, the startup’s higher mobility created enthusiasm across China despite sparking a meltdown within the United States.
DeepSeek’s successful challenge of OpenAI at a reduced financial cost raises doubts regarding the profit potential for US AI leaders such as Microsoft and Nvidia.
It surpassed ChatGPT in Apple’s App Store download numbers, thereby triggering an international market decline in technology shares on Monday.
