China withholds AI test results at 3.6 percent rate

The article discusses the increasing number of tests conducted by Chinese companies on AI models, amid a significant lack of disclosure regarding the safety and security results of these tests. The study revealed that only 3.6% of Chinese models exported between 2021 and 2023 have undergone testing, and less than 1% of them were tested before being released to the market. These findings align with the Chinese government's approach, which focuses on innovation and control, while ignoring Western warnings about the risks of artificial intelligence. Moreover, addressing issues related to the models is limited to simple measures such as cutting power or blocking internet access, as Chinese culture continues to view AI as part of nature rather than a threat. **Summary:** Chinese companies are downplaying the release of AI safety test results, raising concerns about safety and indicating that the government is not enforcing strict oversight. Meanwhile, the benefits and risks of this technology are only gradually slipping into society.

articleDetails.fontSize.sizeحجم النص

تنويه: هذا ملخص تم إنشاؤه بواسطة الذكاء الاصطناعي

عرض المقال بالكامل