SubscribeLog in
Connect with us

China Digital

China’s 2024 Gaokao Triggers Online Discussions on AI

It’s Gaokao time! For the first time, China’s Gaokao essay topic was about the latest AI developments, triggering discussions on social media.

Manya Koetse

Published

on

This week, China’s National College Entrance Exams, better known as the “Gaokao” (高考), became one of the most-discussed topics on Chinese social media. ‘Gaokao,’ ‘AI,’ and ‘Gaokao essay’ were the hottest words on Weibo by the end of the week.

The Gaokao (literally: ‘higher exams’) are a prerequisite for entering China’s higher education institutions and are usually taken by students in their last year of senior high school. June 7th marked the first day of the Gaokao, which will continue until June 9th.

For the over 13.4 million participating students, the Gaokao week is a pivotal moment. Scoring high on this exam can grant access to better colleges, significantly improving their chances of obtaining a good job after graduation. Given the potentially life-changing results, the Gaokao period is a stressful time for both students and their parents.

The Gaokao essay (高考作文) is a significant component of the Chinese language exam, testing students’ writing skills, critical thinking, and ability to express ideas coherently. The essay, which must be completed within a limited time, requires students to discuss given topics.

These topics are generally related to Chinese society and culture, consistently attracting attention on social media. This year, multiple essay questions were related to AI and social media.

Those taking the Beijing exam (北京卷), for example, received a question related to the “like” function on WeChat, suggesting that some people feel strongly about the number of “likes” they receive and give, asking students to reflect on the phenomenon of receiving and giving “likes” on social media.

But the question receiving the most attention on social media was part of the New Curriculum Standard Test I (新课标I卷), which is distributed among different provinces.

 
Students vs. Chatbots: Letting AI Write an Essay on AI
 

Students received the following topic prompt for their Gaokao essay, which should be at least 800 characters long:

With the spread of the internet and AI applications, we can quickly get answers to more and more questions. Will this also lead to us having fewer problems?” (随着互联网的普及、人工智能的应用,越来越多的问题能很快得到答案。那么,我们的问题是否会越来越少?)

The question sparked discussions because it was the first time a Gaokao essay question focused on AI applications designed to interact with users, like ChatGPT.

Although many thought the essay question was easy—unlike this year’s math exam—it still generated some interesting reflections.

Some Weibo users responded that the answer to the question was within the question itself. One Weibo blogger answered: “If there were no AI, we wouldn’t have this question, so problems/questions related to AI will only increase. The emergence of new things will inevitably be accompanied by new problems.”

Others commented on the concerns brought by the emergence of AI applications like ChatGPT. In early 2023, hashtags such as “Ten Professions That Could be Replaced by ChatGPT” (#可能被ChatGPT取代的10大职业#) gained a lot of attention on Chinese social media, where many were concerned that jobs from various industries, including customer service, programming, media, education, market research, finance, etc., would soon be done by AI chatbots instead of humans.

In line with this thought and the AI theme of this year’s Gaokao essay, netizens and some Chinese media outlets decided to let various language models take the exam and see how they would perform: could AI actually do better in writing an essay about AI than Chinese students? (Related hashtags: #让AI写高考作文#, #假如AI写高考语文作文#; #ChatGPT来交卷了#).

Xinhua, for example, asked Open AI’s ChatGPT to answer various questions. Sina Tech (新浪科技) also asked OpenAI’s ChatGPT to answer the AI-related essay question, along with Chinese chatbots, like Baidu’s Erniebot (文心一言), Tencent’s Yuanbao (元宝), iFlytek’s Sparkdesk (星火), and Alibaba’s Tongyi Qianwen (通义千问). Toutiao News did the same, letting three chatbots (ChatGPT, Erniebot, Tongyi Qianwen) compete against each other.

There is no general consensus on how well these AI chatbots can write a Gaokao essay. Although compared to last year, there are now more and better domestic AI models, OpenAI’s ChatGPT-4 is generally still seen as the best chatbot out there.

This was its answer (originally in Chinese, translated to English with ChatGPT):

 

New Issues and Thoughts Brought by Technological Progress

With the popularization of the internet and the application of artificial intelligence, we can quickly obtain a large amount of information, and many questions can be answered in a short time. However, does this mean that we will have fewer questions? I believe the answer is no.

Firstly, although technology provides us with convenience, new technologies also bring new problems. The popularization of the internet has made information dissemination faster, but at the same time, the authenticity and accuracy of information have become new concerns. Fake news, information overload, and privacy breaches are increasingly prominent issues. These new problems need to be identified, understood, and resolved by us.

Secondly, the application of artificial intelligence, although it improves efficiency, also brings ethical and social issues. The transparency of AI decision-making, algorithmic bias, and the impact on the job market are all urgent issues that society needs to address. People need to constantly raise new questions to ensure a balance between technological progress and social welfare.

Furthermore, the development of technology prompts us to reflect more deeply on humanity itself. We need to rethink the relationship between humans and technology and consider the role and boundaries of technology in our lives. These philosophical issues are more important than ever.

Therefore, technological progress does not reduce our questions but rather prompts more new questions and reflections. It is these constantly emerging questions that drive social progress and human thought. We need to embrace the challenges brought by technology with a more open and critical attitude and actively seek solutions.

 

Although the domestic chatbots gave somewhat different answers, the overall tone was similar, though some commenters argued that ChatGPT was still the strongest, along with iFlytek’s Sparkdesk.

An online poll asking Weibo users to grade the ChatGPT essay from lower than 20 points up to the full 60 points saw divided responses, though a majority rated it as lower than 20 points.

How well can ChatGPT write an essay about AI? Opinions are divided.

This shows that many commenters think that AI chatbots are still not able to beat humans when it comes to writing Gaokao essays.

Commenters reacted to the various AI-generated essays in various ways, including:

• “Actually, none of them are very good. They are too formulaic and standardized, lacking the natural creativity and originality that humans possess.”
• “They just give soulless standard answers.”
• “It’s all about ‘firstly,’ ‘secondly,’ ‘furthermore.'”
• “There are no examples, no points proven; it should be a low grade.”
• “It’s just too stiff.”
• “This is like reading reports, not essays.”
• “AI places more emphasis on logic, which aligns with the writing style of foreigners.”
• “There’s no feeling in these essays; there’s a certain kind of AI feeling to AI.”

Meanwhile, some bloggers are taking up the challenge and are publishing their own online essays in response to the Gaokao question.

Some of them are not worried that chatbots will take over their critical tasks: “AI will be AI. There’s no connection to the social realities, and it’s as cold as ice.”

“Their words might make sense, but they lack feeling.”

But for some discussing the topic, they have come to realize that they are already depending too much on digital tools and AI applications for their everyday tasks, writing: “I made an attempt to write an essay, but discovered I already forgot how to do it!” For them, the discussion itself is a wake-up call that writing an essay from scratch is a skill that requires practice and cannot be fully replaced by chatbots, making personal creativity essential to score points and avoid the ‘AI-fication’ of texts.

PS:
In his book China’s Millennials, Eric Fish describes the limits on Chinese students’ answers; taboo responses, such as those containing harsh criticisms of the Chinese government or society, could potentially lead to failure. Although the essay is purportedly meant to showcase the student’s creativity, it must adhere to the unwritten rules of what is socially acceptable.

By Manya Koetse

Spotted a mistake or want to add something? Please let us know in comments below or email us. First-time commenters, please be patient – we will have to manually approve your comment before it appears.

©2024 Whatsonweibo. All rights reserved. Do not reproduce our content without permission – you can contact us at info@whatsonweibo.com.

Manya is the founder and editor-in-chief of What's on Weibo, offering independent analysis of social trends, online media, and digital culture in China for over a decade. Subscribe to gain access to content, including the Weibo Watch newsletter, which provides deeper insights into the China trends that matter. More about Manya at manyakoetse.com or follow on X.

Continue Reading
Click to comment

Leave a Reply

Your email address will not be published. Required fields are marked *

China Arts & Entertainment

How K-pop Fans and the 13-Year-Old Daughter of Baidu VP Sparked a Debate on Online Privacy

What began as K-pop fan outrage targeting a snarky commenter quickly escalated into a Baidu-linked scandal and a broader conversation about data privacy on Chinese social media.

Ruixin Zhang

Published

on

For an ordinary person with just a few followers, a Weibo account can sometimes be like a refuge from real life—almost like a private space on a public platform—where, along with millions of others, they can express dissatisfaction about daily annoyances or vent frustration about personal life situations.

But over recent years, even the most ordinary social media users could become victims of “opening the box” (开盒 kāihé)—the Chinese internet term for doxxing, meaning the deliberate leaking of personal information to expose or harass someone online.

 
A K-pop Fan-Led Online Witch Hunt
 

On March 12, a Chinese social media account focusing on K-pop content, Yuanqi Taopu Xuanshou (@元气桃浦选手), posted about Jang Wonyoung, a popular member of the Korean girl group IVE. As the South Korean singer and model attended Paris Fashion Week and then flew back the same day, the account suggested she was on a “crazy schedule.”

In the comment section, one female Weibo user nicknamed “Charihe” replied:

💬 “It’s a 12-hour flight and it’s not like she’s flying the plane herself. Isn’t sleeping in business class considered resting? Who says she can’t rest? What are you actually talking about by calling this a ‘crazy schedule’..

Although the comment may have come across as a bit snarky, it was generally lighthearted and harmless. Yet unexpectedly, it brought disaster upon her.

That very evening, the woman nicknamed Charihe was bombarded with direct messages filled with insults from fans of Jang Wonyoung and IVE.

Ironically, Charihe’s profile showed she was anything but a hater of the pop star—her Weibo page included multiple posts praising Wonyoung’s beauty and charm. But that context was ignored by overzealous fans, who combed through her social media accounts looking for other posts to criticize, framing her as a terrible person.

After discovering through Charihe’s account that she was pregnant, Jang Wonyoung’s fans escalated their attacks by targeting her unborn child with insults.

The harassment did not stop there. Around midnight, fans doxxed Charihe, exposing her personal information, workplace, and the contact details of her family and friends. Her friends were flooded with messages, and some were even targeted at their workplaces.

Then, they tracked down Charihe’s husband’s WeChat account, sent him screenshots of her posts, and encouraged him to “physically punish” her.

The extremity of the online harassment finally drew backlash from netizens, who expressed concern for this ordinary pregnant woman’s situation:

💬 “Her entire life was exposed to people she never wanted to know about.”

💬 “Suffering this kind of attack during pregnancy is truly an undeserved disaster.

Despite condemnation of the hate, some extreme self-proclaimed “fans” remained relentless in the online witch hunt against Charihe.

 
Baidu Takes a Hit After VP’s 13-Year-Old Daughter Is Exposed
 

One female fan, nicknamed “YourEyes” (@你的眼眸是世界上最小的湖泊), soon started doxxing commenters who had defended her. The speed and efficiency of these attacks left many stunned at just how easy it apparently is to trace social media users and doxx them.

Digging into old Weibo posts from the “YourEyes” account, people found she had repeatedly doxxed people on social media since last year, using various alt accounts.

She had previously also shared information claiming to study in Canada and boasted about her father’s monthly salary of 220,000 RMB (approx. $30.3K), along with a photo of a confirmation document.

Piecing together the clues, online sleuths finally identified her as the daughter of Xie Guangjun (谢广军), Vice President of Baidu.

From an online hate campaign against an innocent, snarky commenter, the case then became a headline in Chinese state media, and even made international headlines, after it was confirmed that the user “YourEyes”—who had been so quick to dig up others’ personal details—was in fact the 13-year-old daughter of Xie Guangjun, vice president at one of China’s biggest tech giants.

On March 17, Xie Guangjun posted the following apology to his WeChat Moments:

💬 “Recently, my 13-year-old daughter got into an online dispute. Losing control of her emotions, she published other people’s private information from overseas social platforms onto her own account. This led to her own personal information also getting exposed, triggering widespread negative discussion.

As her father, I failed to detect the problem in time and failed to guide her in how to properly handle the situation. I did not teach her the importance of respecting and protecting the privacy of others and of herself, for which I feel deep regret.

In response to this incident, I have communicated with my daughter and sternly criticized her actions. I hereby sincerely apologize to all friends affected.

As a minor, my daughter’s emotional and cognitive maturity is still developing. In a moment of impulsiveness, she made a wrong decision that hurt others and, at the same time, found herself caught in a storm of controversy that has subjected her to pressure and distress far beyond her age.

Here, I respectfully ask everyone to stop spreading related content and to give her the opportunity to correct her mistakes and grow.

Once again, I extend my apologies, and I sincerely thank everyone for your understanding and kindness.

The public response to Xie’s apology has been largely negative. Many criticized the fact that it was posted privately on WeChat Moments rather than shared on a public platform like Weibo. Some dismissed the statement as an attempt to pacify Baidu shareholders and colleagues rather than take real accountability.

Netizens also pointed out that the apology avoided addressing the core issue of doxxing. Concerns were raised about whether Xie’s position at Baidu—and potential access to sensitive information—may have helped his daughter acquire the data she used to doxx others.

Adding fuel to the speculation were past conversations allegedly involving one of @YourEyes’ alt accounts. In one exchange, when asked “Who are you doxxing next?” she replied, “My parents provided the info,” with a friend adding, “The Baidu database can doxx your entire family.”

Following an internal investigation, Baidu’s head of security, Chen Yang (陈洋), stated on the company’s internal forum that Xie Guangjun’s daughter did not obtain data from Baidu but from “overseas sources.”

However, this clarification did little to reassure the public—and Baidu’s reputation has taken a hit. The company has faced prior scandals, most notably a the 2016 controversy over profiting from misleading medical advertisements.

 
Online Vulnerability
 

Beyond Baidu’s involvement, the incident reignited wider concerns about online privacy in China. “Even if it didn’t come from Baidu,” one user wrote, “the fact that a 13-year-old can access such personal information about strangers is terrifying.”

Using the hashtag “Reporter buys own confidential data” (#记者买到了自己的秘密#), Chinese media outlet Southern Metropolis Daily (@南方都市报) recently reported that China’s gray market for personal data has grown significantly. For just 300 RMB ($41), their journalist was able to purchase their own household registration data.

Further investigation uncovered underground networks that claim to cooperate with police, offering a “70-30 profit split” on data transactions.

These illegal data practices are not just connected to doxxing but also to widespread online fraud.

In response, some netizens have begun sharing guides on how to protect oneself from doxxing. For example, they recommend people disable phone number search on apps like WeChat and Alipay, hide their real name in settings, and avoid adding strangers, especially if they are active in fan communities.

Amid the chaos, K-pop fan wars continue to rage online. But some voices—such as influencer Jingzai (@一个特别虚荣的人)—have pointed out that the real issue isn’t fandom, but the deeper problem of data security.

💬 “You should question Baidu, question the telecom giants, question the government, and only then, fight over which fan group started this.

As for ‘Charihe,’ whose comment sparked it all—her account is now gone. Her username has become a hashtag. For some, it’s still a target for online abuse. For others, it is a reminder of just how vulnerable every user is in a world where digital privacy is far from guaranteed.

By Ruixin Zhang

Independently covering digital China for over a decade. Like what we do? Support us and get the story behind the hashtag by subscribing:

edited for clarity by Manya Koetse

Spotted a mistake or want to add something? Please let us know in comments below or email us. First-time commenters, please be patient – we will have to manually approve your comment before it appears.

©2024 Whatsonweibo. All rights reserved. Do not reproduce our content without permission – you can contact us at info@whatsonweibo.com.

Continue Reading

China Digital

Chinese Netizens Turn to Tim Cook Over Battery Factory’s Illegal Overtime

Is turning to Western suppliers an effective way for workers to pressure domestic companies into complying with labor laws?

Manya Koetse

Published

on

🇨🇳 CHINA TRENDING WEEK 12: WHAT’S ON WEIBO SHORTS
We include this content in the Weibo Watch newsletter. Subscribe to get it in your inbox 📩

Recently, Chinese netizens have started reaching out to Apple and its CEO Tim Cook in order to put pressure on a state-owned battery factory accused of violating labor laws.

The controversy involves the Huizhou factory of Desay Battery (德赛电池), known for producing lithium batteries for the high-end smartphone market, including Apple and Samsung. The factory caught netizens’ attention after a worker exposed in a video that his superiors were deducting three days of wages because he worked an 8-hour shift instead of the company’s “mandatory 10-hour on-duty.” Compulsory overtime violates China’s labor laws.

In response, the worker and other netizens started to let Apple know about the situation through email and social media, trying to put pressure on the factory by highlighting its position in the Apple supply chain. In at least one instance, Apple confirmed receipt of the complaint. (Meanwhile, on Tim Cook’s official Weibo account, the comment section underneath his most recent post is clearly being censored.)

Screenshot of replies on X underneath a post by Tim Cook on International Women’s Day.

The factory, however, has denied the allegations, , claiming that the video creator was spreading untruths and that they had reported him to authorities. His content has since also been removed. A staff member at Desay Battery maintained that they adhere to the 8-hour workday and appropriately compensate workers for overtime.

At the same time, Desay Battery issued an official statement, admitting to “management oversights regarding employee rights protection” (“保障员工权益的管理上存在疏漏”) and promising to do better in safeguarding employee rights.

One NetEase account (大风文字) suggested that for Chinese workers to effectively expose labor violations, reporting them to Western suppliers or EU regulators is an effective way to force domestic companies to respect labor laws.

Another commentary channel (上峰视点) was less optimistic about the effectiveness, arguing that companies like Apple would be quick to drop suppliers over product quality issues but more willing to turn a blind eye to labor violations—since cheap labor remains a key competitive advantage in Chinese manufacturing.

By Manya Koetse

(follow on X, LinkedIn, or Instagram)

 

Spotted a mistake or want to add something? Please let us know in comments below or email us. First-time commenters, please be patient – we will have to manually approve your comment before it appears.

©2025 Whatsonweibo. All rights reserved. Do not reproduce our content without permission – you can contact us at info@whatsonweibo.com.

Follow What’s on Weibo on

Continue Reading

Subscribe

What’s on Weibo is run by Manya Koetse (@manyapan), offering independent analysis of social trends in China for over a decade. Subscribe to gain access to all content and get the Weibo Watch newsletter.

Manya Koetse's Profile Picture

Get in touch

Would you like to become a contributor, or do you have any tips or suggestions? Get in touch here!

Popular Reads