Deepseek suffers cyber attack as users migrate for Chinese AI startup

(Reuters)-Chinese startup said on Monday that it will temporarily limit entries due to a cyber attack after the company’s artificial intelligence assistant gain sudden popularity.

At the beginning of the day, the startup was also hit by service interruptions on its website after its AI assistant became the United States.

The company has solved problems related to the application programming interface and the inability of users to login on the site, according to their status page. The service interruptions on Monday were the longest of the company in about 90 days and coincide with its high popularity.

Last week, Deepseek launched a free assistant that, according to the company, uses less data for a fraction of the cost of the established models, which can mark a turn point at the investment level required for AI.

Read more:

Fueled by the Deepseek-V3 model, which, according to its creators, “is at the top of the table between open source models and rivals the most advanced closed code models in the world”, the AI ​​application has grown in popularity among US users Since its launch on January 10, according to the Tower Sensor Application Data Research Company.

This milestone highlights how Deepseek has left a deep impression on Silicon Valley, overthrowing widespread views on USA AI primacy and the effectiveness of Washington export controls aiming at advanced chip and AI from China.

Technology actions were harmed on Monday, with the roles of Nvidia and Oracle plummeting.

AI models from Chatgpt to DeepSeek require advanced chips to feed their training. Since 2021, the Biden government has expanded the scope of prohibitions designed to prevent these chips from being exported to China and used to train AI models of Chinese companies.

However, DeepSek researchers wrote in an article last month that DeepSeek-V3 used Nvidia’s H800 chips for training, spending less than $ 6 million.

Although this detail has been contested since then, the claim that used chips were less powerful than the most advanced Nvidia products that Washington has sought to keep out of China, as well as relatively cheap training costs, took US technology executives from the US technology executives questioning the effectiveness of technology export controls.

Little is known about the company behind DeepSeek, a small Hangzhou -based startup founded in 2023, when Baidu, a giant of search engines, launched the first Chinese model of large AI -scale language.

Since then, dozens of big and small Chinese technology companies have launched their own AI models, but DeepSeek is the first to be praised by the US technology industry for matching or even overcoming the performance of north-tipped models- Americans.