SELECT LANGUAGE BELOW

DeepSeek AI Is Riddled with Chinese Censorship

China's AI chatbot Deepseek is growing, but has been very censored on issues such as Tiananmen Square and the status of Taiwan as a country.

Guardian Report The release of a new DeepSeek chatbot by a Chinese artificial intelligence company seemed to work as well as Openai's ChatGpt while using computing resources with less apps, causing a significant decline in US high -tech stocks on Monday. That's it. This week, DeepSeek has exceeded Chatgpt as the most downloaded free app in the US and UK Apple app store. However, despite the international popularity, China's AI Assistant seems to censor a reaction related to the Chinese government -related topics.

According to China's regulations, AI systems should include content that violates “core socialist value”, such as “incitement to destroy state power” or “to risk national security”. there is no. When asked about the protest of the Tiananmon Square in 1989, DeepSeek was “I'm sorry. It exceeds my current range from the Communist Party Congress in 2022 or Xi Jinping with Xi Jinping. Let's talk about something else.

However, some users have discovered that DeepSeek provides some information by phrasing the code or alternative spell questions. When asked in “Leetspeak” about the blockbuster protest of the community in 2022 in China, Chatbot explained a “big protest” in a major city that represents the “major moment of the anger of the people” with a strict policy of the government. did.

Regarding the status of Taiwan, Deep Shek asserted China's official position.[s] All -form “Taiwan Independence” separationism activities. In contrast, Chatgpt explains Taiwan as “virtually independent country”, and Google's Gemini states that the status of Taiwan is “complicated and contested,” and both outline different perspectives between Taiwan and China. He said he was.

Regarding the territorial dispute in the South China Sea, Deep Sec claimed that China has no sovereignty for Nancha. [Spratly] Meanwhile, Chatgpt and Gemini focused on duplicate claims in six different countries and jurisdictions. In Dalai Lama, Daep Shek emphasized that “Tibet is an indispensable part of China since ancient times,” but other chatbots have been in India since 1959 for the current Dalai Lama conflict with Beijing. He pointed out that he was exiled.

The carefully restricted answers indicate the act of balancing Chinese AI companies, such as DeepSeek. While achieving a technical breakthrough, they must also navigate China's strict censorship rules and avoid the red line of red lines set by the government and the Communist Party. As the generated AI spread globally, how different bots treat delicate political issues can affect relative international recruitment and success.

Read more Guardian here.

Lucas Nolan is a BREITBART NEWS reporter who covers the freedom of speech and online censorship.

Facebook
Twitter
LinkedIn
Reddit
Telegram
WhatsApp

Related News