What happens outside the exam hall is often more significant than what happens inside.
Article author and source: Beyond the page
The 2026 college entrance examination saw an unprecedented phenomenon.
Even before the test takers received their scores, AI had already started vying for their spots.
Qianwen launched its College Entrance Examination Application Agent, touted as China's first intelligent agent for full-cycle college entrance examination application, and it's free. Quark upgraded its College Entrance Examination channel, opening up all four major functions, serving college entrance examination users for the eighth consecutive year.
Tencent's Yuanbao, in partnership with QQ Browser, launched Yuanbao Gaokao Tong, positioning itself as a Gaokao (college entrance exam) consultant agent. Baidu directly embedded its Gaokao module into Wenxin Assistant. Although Doubao doesn't have a separate section, its chat box can already answer most questions about college application.
Almost all mainstream AI products are vying for the same entry point: college application.
On the surface, this is just a marketing campaign capitalizing on a trending topic. Every year during the college entrance examination season, various internet companies vie for traffic by targeting students and parents.
However, if we extend the timeline a bit, we'll discover something even more noteworthy.
For many years, the biggest business related to the college entrance examination has been information asymmetry. But in 2026, for the first time, information asymmetry will begin to become a public good.
This marks the first time AI has entered the life decision-making process of Chinese people on a large scale. It is also the first time AI has begun to influence the future choices of tens of millions of families behind 12.9 million test takers.
The real change in this year's college entrance examination didn't happen inside the examination hall; it happened outside.
1. Something that used to cost 10,000 yuan is suddenly free.
The business of college application guidance has been expanding over the past decade.
From early supplementary textbooks and score line handbooks to later college application consultants, one-on-one consultations, and college admission agencies, the fear of parents choosing the wrong major has reached its peak amid the transformation of the education and training industry and the fierce competition for civil service exams, directly giving rise to a distorted and booming college application market.
High-end college application services on the market today have already exceeded 10,000 yuan. One-on-one consultations often cost 12,999 to 18,999 yuan, while a 40-minute consultation costs 5,000 yuan. Places are sold out quickly, and popular consultants are booked up for up to three years in advance. Regular consultants cost 5,000 to 8,000 yuan. Newly opened similar agencies in county towns start at 3,000 to 5,000 yuan.
What are they selling? On the surface, it's consulting services, but in essence, it's information asymmetry.
They don't create admission slots, nor can they help students score 50 more points. What they create is a cognitive advantage: which schools might lower their admission scores this year, which majors offer better future employment prospects, which cities are more worthwhile to visit, and which combination of application choices will best ensure that scores are not wasted.
This information is not a secret.
But for an 18-year-old student who has never experienced college application before and his parents, it is almost impossible to complete the collection, screening, judgment and decision-making in a few days. Thus, an industry based on information asymmetry was born.
Today, AI is doing the same thing. Only it's free.
Last year, Alibaba pioneered the "AI College Application Report" format, distributing nearly 13 million AI-powered college application reports. This year, the report has been upgraded, with each report ranging from 15 to 40 pages, covering dozens of college application combinations. Students input their province, chosen subjects, scores, and rankings, and the system generates ambitious, realistic, and safe options, along with admission trends, major analysis, and employment advice.
Two or three years ago, these things would cost many families tens of thousands of yuan. Today, all you need to do is open an app.
Second, AI is not eliminating planners.
The impact of technology on the traditional consulting industry is far more rapid than imagined.
In the past: You didn't understand the rules, but I did. So you had to pay.
Today: AI learns the rules faster than anyone else, and for free.
Yuanbao Gaokaotong accesses a college admission database spanning over a decade. Its Qianwen agent can identify a student's interests, city preferences, career aspirations, and even MBTI personality traits. Even two students from the same province with the same score and subject selection will receive completely different recommendations. It claims to cover nearly 3,000 universities and over 2,000 undergraduate majors, with data verified through multi-source cross-validation.
They may not be perfect. But for the vast majority of students who never received any professional guidance in the past, China has a large number of first-generation college students from rural areas and county towns every year. Their parents have never left their county towns in their entire lives and are simply unable to guide them in choosing their majors.
For this group of people, the emergence of AI has provided a safety net for basic information.
Families in Beijing, Shanghai, and Shenzhen naturally have access to more educational information. They know which majors are emerging, which schools are undervalued, and which sectors have the greatest potential over the next decade.
Many families in county towns do not have such resources. With the same score, parents in big cities may help their children avoid professions in declining industries, while parents in county towns may let their children miss out on industry opportunities simply because it sounds respectable.
Thus, the same score can lead to completely different life paths. After the college entrance examination, what truly differentiates people is often not just the exam itself, but also the information they receive.
The intervention of AI is essentially using the certainty of technology to combat the uncertainty brought about by region and class. This year, for the first time, a new possibility has emerged: the information gap after the exams are also beginning to be smoothed out by AI.
This may be the closest China's education system has come to information equality in many years.
III. Why has the National College Entrance Examination (Gaokao) become a scenario that AI is vying to conquer?
It's not because the college entrance exam generates a lot of traffic, but because the college entrance exam is naturally suited for AI.
The data is highly standardized. Nearly 3,000 universities nationwide, thousands of majors, and their historical admission scores and rankings are all publicly available and transparent. This is the type of data that AI excels at processing.
The decision-making chain is extremely clear. The inputs are scores, rankings, subject selections, and preferences. The output is a list of university application choices. The logic in between can be fully structured. This is precisely the type of task that agents excel at: given a goal, plan a path, and output a solution.
The demand is extremely rigid. After this year's college entrance examination, 12.9 million students must make a crucial decision that could impact their future for years to come within a very short timeframe. Time is limited, the pressure is immense, and the margin for error is very low. This scenario naturally drives users to seek more efficient tools.
Standardized data, a clear decision-making chain, and strong demand. These three conditions combined make the National College Entrance Examination (Gaokao) one of the easiest real-world application scenarios for AI to successfully implement in China.
Fourth, the real change is not that applying to universities has become easier.
What did a student typically go through when filling out their college application in the past?
Asking teachers is risky because they are only familiar with their own school's past experiences. Asking parents often limits their knowledge to their own era. Asking alumni only provides their own perspectives. Hiring an agency is expensive and the quality varies greatly.
Every source of information is partial, limited, and biased.
Today, something new emerged for the first time. AI may not be perfect, but it has seen all the data and will not recommend a particular school based on vested interests or reject a particular major based on personal preferences. It can complete the comparison of solutions in tens of seconds that previously took days.
This is not just an upgrade of tools, but a reconstruction of cognitive resources.
For the first time, a student from a top-tier high school in Shenzhen and a student from a county-level high school in Guizhou had the opportunity to receive volunteer recommendations of nearly equal quality.
It's not because their scores are the same, but because AI doesn't distinguish who you are; as long as you input your score, the quality of the analysis is the same.
In the past, personal connections determined access to information. Today, accessing AI means accessing information.
V. The First Door for AI to Enter Chinese Homes
Many people believe that writing code, generating images, and generating videos constitute the AI revolution.
However, this is not the case for the vast majority of ordinary Chinese families. They don't write code, make videos, or even have much contact with cutting-edge AI products.
The college entrance examination is different.
Behind the 12.9 million test takers are tens of millions of parents. When AI began to participate in college application, it entered for the first time the most important, sensitive, and error-tolerant decision-making scenario for Chinese families.
Once this door is opened, everything else will fall into place. Today it's choosing a university, tomorrow it might be choosing a job, and the day after tomorrow it might be choosing a house, managing finances, planning a career, or even planning your life.
People place their trust in AI around the time of the college entrance examination. This trust will seep into every major choice they make in the future, transforming AI from a tool into an advisor.
The National College Entrance Examination (Gaokao) may be the starting point of this change.
VI. When life is precisely calculated
AI has access to all historical data and can calculate the employment rate, average salary, and industry lifespan of every major using the most basic return on investment. Its recommendations are therefore inevitably rational, risk-free, and even pragmatic.
In AI algorithms, "surge, stabilize, and protect" is a sophisticated mathematical problem. It might advise a child with literary talent to study computer science or law because the latter offers a wider range of job opportunities. It might suggest that a student obsessed with archaeology choose financial management because the career return cycle for that major is shorter.
At this point, what choices and reflections are left in life?
If we completely follow AI's lead, we may indeed avoid risks and achieve the highest probability of social average success. But are we also giving up something else in the process: blind, enthusiastic, and reckless favoritism, and the freedom to make mistakes?
Information asymmetry has been smoothed out. But will the diversity of human life also be smoothed out by algorithms?
AI can help us avoid misjudgments on college application forms, but it can't take away our burdens of confusion throughout life. Faced with a flawless college application form generated by an algorithm, the one who truly needs to make the decision is still that young person standing at the crossroads of turning 18.
A note outside the main page:
Many years from now, people may not remember what the 2026 college entrance examination essay topic was, nor may they remember how difficult the last math problem was.
But they might remember something else.
That year, AI appeared on a college application form for the first time.
Starting with that college application form, AI participated in the life choices of ordinary Chinese people for the first time.
In the past, people always understood AI as a technology, until it began to affect people's lives, and only then did the technology truly become a reality.
What truly deserves to be remembered is not how many decisions AI has made for us, but whether, in this era of increasingly intelligent algorithms, we are still willing to bear all the consequences of our choices.
