What's left for humans as AI becomes smarter? On the morning of September 10, the opening ceremony and main forum of the 2026 Inclusion · Bund Conference were held in Shanghai. During a roundtable discussion, renowned writer Liu Zhenyun and Professor Ma Yi, Dean of the School of Computing and Data Science at the University of Hong Kong, discussed the changes brought by AI in creation, education, employment, and human relationships from the different perspectives of literature and technology.

Both of their assessments are not pessimistic. Liu Zhenyun said that AI is "not a monster"; Ma Yi reminded that the real issue is not whether people should use AI, but how people should re-understand learning and their own value.

1789030617946.jpg

AI Takes Away Commonality, Humans Hold Onto Individuality

How far has AI come? Liu Zhenyun has a direct experience: a large number of videos online featuring his image and voice are "95% fake." AI imitates his voice and lip movements, using works such as "One Heap of Chickens" and "One Sentence Worth Ten Thousand" to continue "expressing his views."

He compares this to Sun Wukong and the Monkey King Six Ears — "Even the Buddha couldn't tell which one was real." Facing the "AI Liu Zhenyun," his feeling is not fear, but "very humorous."

But in Liu Zhenyun's view, being similar does not mean true creation. AI is good at handling past and current knowledge, but it is difficult for it to create new perspectives. "It can imitate 'One Heap of Chickens' to write 'One Heap of Geese,' but I haven't thought of or written a work that AI could not imitate."

In Ma Yi's view, this exactly reveals a boundary of large models. AI can now grasp knowledge in literature, history, and science very well, even reaching professional levels, but these are largely the "commonalities" accumulated by humans. "Everyone has their own individuality; our joys, sorrows, and values are all very different," Ma Yi said. AI can explain common things very well, but it is difficult to understand what an individual values most.

The real more difficult question then arises: where is the value of humans in the face of technological change? "This is a question that everyone must think about now," Ma Yi said. "A better AI era is when each person can find a new balance between the common values of human society and personal value."

"AI Is Not a Monster"

Anxiety about AI has already extended to education and employment. Ma Yi said that many of his students and parents are anxious: What major should they study so that they won't be replaced by AI in the future? But in his view, trying to find a skill that will never be replaced is itself an old way of thinking.

Some changes have already occurred. Ma Yi believes that in clearly defined tasks such as programming and mathematical reasoning, AI can now be greatly automated, even replacing some jobs. But this doesn't mean that computer science or mathematics lose their value. On the contrary, the more important thing in the future is to use computer science and mathematics as tools to solve scientific, engineering, and social problems.

Therefore, in the AI general education course at the University of Hong Kong, he encourages students to fully embrace AI. He says that instead of looking for a skill that AI can never learn, it's better to learn how to use AI to further amplify the abilities they already have.

Liu Zhenyun used the mechanization of rural areas to respond to concerns about unemployment. In the past, farming relied on human labor and animals, but later there were tractors and harvesters. Traditional labor decreased, but people did not lose their livelihoods. "Now, farmers in my village can farm hundreds of acres, but do the others become unemployed? No."

Liu Zhenyun said that the flow of productivity creates new production relations, and the benefits of technology outweigh the costs. "AI is definitely not a monster."

AI Can Recommend Restaurants, But Won't Invite You to Dinner

A deeper change than work is human relationships. More and more people turn to AI for emotional support. Can AI ultimately solve human loneliness?

Liu Zhenyun believes there is a clear boundary between knowledge acquisition and emotional communication. AI can answer life questions, but two friends talking under the lamp is still different from chatting with AI — "It is common, while you are unique."

He gave an example that made the audience laugh: Going to Hong Kong, he can ask AI which restaurant is good; but he would directly ask Ma Yi: "When will you treat me to dinner?" Liu Zhenyun continued, "AI will only tell you which restaurant is good, but it won't invite you to dinner. Professor Ma will treat me to dinner."

Ma Yi also reminded of the other side of the coin: AI can help students who are afraid to ask questions in public to learn more freely. However, if knowledge acquisition shifts entirely to AI, students may become more isolated.

"In the end, people must enter society and interact with real people," Ma Yi said. There needs to be a new balance between the convenience of AI and human social skills.