DeepSeek Hit by Cyberattack as AI Popularity Soars

News Desk

Islamabad: Chinese startup DeepSeek announced on Monday that it will temporarily limit new registrations due to a cyberattack, following a surge in popularity of its AI assistant.

Earlier in the day, the company also experienced outages on its website after the assistant became the top-rated free app on the U.S. Apple App Store. DeepSeek resolved issues related to its application programming interface and login problems, according to its status page.

The outages on Monday marked the company’s longest downtime in about three months, coinciding with its rapid rise in popularity.

Last week, DeepSeek launched its free assistant, which it claims uses significantly less data and is much more cost-effective compared to competing AI models. Powered by the DeepSeek-V3 model, which its creators assert is a leader among open-source models and rivals top closed-source models worldwide, the application quickly gained traction with U.S. users since its January 10 release, according to Sensor Tower.

This development highlights DeepSeek’s growing influence in Silicon Valley, challenging the prevailing belief in U.S. dominance in AI and questioning the effectiveness of Washington’s export controls targeting China’s advanced chip and AI capabilities.

Monday’s technology stock sell-off saw shares of Nvidia and Oracle take a significant hit. AI models like ChatGPT and DeepSeek rely on advanced chips for training. Since 2021, the Biden administration has expanded export bans to prevent the shipment of high-end chips to China for use in AI model training.

However, DeepSeek researchers revealed in a paper last month that their DeepSeek-V3 model used Nvidia’s H800 chips for training, costing under $6 million.

While this claim has been disputed, the suggestion that the chips were less powerful than the most advanced Nvidia products targeted by U.S. export controls, combined with the relatively low training costs, has led U.S. tech executives to question the efficacy of these restrictions.

Little is known about DeepSeek, a small startup based in Hangzhou, founded in 2023, the same year Baidu launched China’s first AI large-language model.

Since then, numerous Chinese companies have introduced their own AI models, but DeepSeek stands out as the first to receive recognition from the U.S. tech industry for matching or even surpassing the performance of leading U.S. models.