China Implements AI Chatbot Restrictions During College Entrance Exam Period
Photo Recognition and Explanation Features Blocked by Time Slot
Stricter Crackdown on Cheating This Year

Chinese artificial intelligence (AI) companies have decided to simultaneously block photo-based question-solving features during the national college entrance exam, known as Gaokao. This measure is intended to prevent cheating using advanced technology.


A photo showing the Chinese national flag fluttering to aid understanding of the article. Photo by AP Yonhap News Agency

A photo showing the Chinese national flag fluttering to aid understanding of the article. Photo by AP Yonhap News Agency

View original image

According to Yonhap News, on May 29, China Central Television (CCTV) reported that "major AI platforms will apply special controls to features related to the Gaokao exam, which takes place from the 7th to the 10th." CCTV contacted the main AI platforms, and all companies stated, "The platforms will remain available during the Gaokao period, but special controls will be applied to features related to the exam situation."


This feature restriction will be implemented according to specific time slots. The photo recognition and explanation features will be temporarily blocked only during the hours when the Gaokao is being held, while Q&A functions related to daily life will remain accessible without restriction. In addition to AI platforms, online learning tools that help with homework for elementary, middle, and high school students will also restrict their Q&A functions during the exam period.


The Gaokao, often referred to as "China's college entrance exam," is effectively the sole gateway to higher education and serves as a critical channel for upward mobility, especially for students from rural and low-income families. Last year, approximately 13.35 million candidates took the exam. Electronic devices such as mobile phones are already banned inside the test centers, and authorities have deployed advanced monitoring measures such as signal jammers and AI-based abnormal behavior detection systems.


Chinese IT company ByteDance. Reuters Yonhap News

Chinese IT company ByteDance. Reuters Yonhap News

View original image

Last month, China’s Ministry of Education, the Cyberspace Administration, the Ministry of Public Security, and other government agencies convened the "2026 Gaokao Safety Work Video Conference," announcing plans to conduct exam operations to higher standards and to crack down on cheating using advanced technologies.


This strong regulatory stance sparked debate online in China. Some netizens questioned whether restricting AI functions was necessary, given that exam security is already extremely strict. On the other hand, many argued that controls are inevitable to ensure fairness, as traditional surveillance methods have reached their limits due to the surreptitious use of smart devices and the rapid response speed of AI.



The restriction of AI features during exams was fully implemented last year. Last year, major AI chatbots such as Alibaba’s Qwen, Tencent’s Yuanbao, ByteDance’s Doubao, and Moonshot’s Kimi disabled their photo recognition and Q&A functions during exam hours. At that time, when a student uploaded an exam paper photo to Doubao, the AI chatbot displayed a message stating, "In order to ensure the fairness of the college entrance exam, the Q&A feature will be temporarily suspended during the exam period and will be restored after the test concludes."


This content was produced with the assistance of AI translation services.

© The Asia Business Daily(www.asiae.co.kr). All rights reserved.

Today’s Briefing