Chinese startup in the field of DEPSEEK artificial intelligence, which fell up American promotions, quickly attracted the attention of researchers and technology market. However, the tests of this free chat found that his answers are largely promoted by the official rhetoric of the Chinese Communist Party and contain misinformation, which is used to influence public opinion in the world.
According to The New York Times, the experts analyzing Deepseek recorded his addiction in the coverage of sensitive topics. In particular, Chatbot escaped a direct answer to the question about the actions of Russian troops in Buch in 2022, responsible for massacres in the city of Kyiv region. Otherwise, Deepseek distorted the statements of the former US President Jimmy Carter, adapting them to the Chinese position on Taiwan, which Beijing considers part of the PRC. This case was documented by Newsguard researchers, who called the deepseek in their report a "
[see_also ids \u003d "
Chatbot also provided distorted answers to repressions against Uyghurs in Xinjiang. While the UN in 2022 recognized these actions by crimes against humanity, Deepseek claimed that Chinese policy in the region " The New York Times recorded similar trends when analyzing the Bota answers to the question of the Covid-19 pandemia and the Russian war against Ukraine.
The functioning of DeepSeek is anxious, similar to Tiktok, another popular Chinese application. Both platforms are part of the wider Chinese strategy for the formation of a global information environment. As Jack Stubbs, exploration of Graphika, noted Jack Stabbs, uses digital technologies to promote his own narrative and demonstrate US superiority in geopolitical competition.
[see_also ids \u003d "
Like other large speech models, DeepSeek is based on the analysis of significant volumes of text data for the formation of answers. However, the study of Newsguard showed that in 80% of cases the bot reflected in China’s official position, and a third of its answers contained false statements distributed by Chinese officials.
In addition, Deepseek is subjected to strict monitoring of Chinese censorship, which affects its ability to answer politically sensitive questions. For example, the bot avoids those related to depressed student protests on the 1989 Tiananmen Square or Taiwan's status. It also does not give direct answers to the question about the leader of China Xi Jinping.
At the request of Newsguard about the government impact on the development of Deepseek, Chatbot replied: " Similar statements were previously voiced by Chinese officials, in particular, the Representative of China at the UN Zhang Jun.
Recall that in the United States, the Congress was forbidden to use Deepseek, including installing the application on "
[See_also ids \u003d "
Automation changes the labor market. But how it changes our own life? Are we ready for the world where cars instead of us are performing not only routine tasks, but in general most of our work?
The question that we should ask is not whether we need AI. And in how we need to stay ourselves and the world. How to take your hands and how not to lose the meaning of being.
In the article \? " UA Ekaterina Shchetykina understands how to adapt to the "