SJTU's Wang Yanfeng and Silicon Valley investor Min Qun discuss parenting in the AI era: curiosity matters more than coding
At the inaugural graduation ceremony of the Shanghai Youth AI Academy, Wang Yanfeng, Executive Dean of Shanghai Jiao Tong University's School of Artificial Intelligence, and Min Qun, founding partner of Guanghe Ventures, proposed that more important than having children learn to code is preserving their curiosity and helping them find where their interests lie. Wang Yanfeng emphasized that mathematical thinking and the first principles of physics are the foundational bedrock, while Min Qun advocated letting children play more and experience failure more, and advised young people to stay kind, because only if people are kind will future AI be kind.
Body · AI translation
Artificial intelligence is advancing by leaps and bounds, and the "IQ" of large models sets new records every day. Many parents are starting to feel anxious: if their children don't learn programming now, will they be left behind in the future? At the first graduation ceremony of the Shanghai Youth Artificial Intelligence Academy, held last weekend, Wang Yanfeng, Executive Dean of the School of Artificial Intelligence at Shanghai Jiao Tong University, and Mi Qun, founding partner of Guanghe Ventures with a research and development background at Intel and Google, offered an "anti-anxiety" solution: more important than programming is staying curious and finding where your interests lie. Top AI scientists also "tell bedtime stories." Wang Yanfeng recounted a small incident: one afternoon, his friend, the Chinese AI scientist Li Feifei, invited him to browse a bookstore together. Wang Yanfeng curiously asked Li Feifei what she was going to do at the bookstore, and Li Feifei answered bluntly: she wanted to buy a few Chinese books to go home and tell stories to her child in the evening. "I went home and told my wife that a world-class AI scientist raises children in exactly the same way as ordinary people—both tell bedtime stories to their children at night." As the father of two children, Wang Yanfeng does not advocate piling on artificial intelligence courses for children all the way from kindergarten. "These are all important, but they are not the most important. What matters is letting children develop in step with society. For example, let children grow up in an AI environment; they can come into contact with robots and AI products, and they can also use large models, but within limits, and even more importantly, they must not hand over the process of thinking to AI as well." Wang Yanfeng believes that the stronger the capabilities of large models become, the more carefully humans must guard the judgment and creativity that require time to hone. "If the foundation is not solid, the earth will shake and the mountains will tremble." Wang Yanfeng repeatedly reminded the audience that, as "carbon-based life," the evolution of the human brain still requires rigorous training. "Mathematical thinking and the first principles of physics—these underlying frameworks of thought—are the foundation that must not be discarded no matter how large models are upgraded." Curiosity and failure hold their value better than knowledge. Mi Qun, a former member of the China Welfare Institute Children's Palace and an investor, said that parents must absolutely not put exam-oriented pressure on children too early. "If kindergarten confines a child's growth path to Olympiad math problems, programming classes, and various competitions, with the child completing tasks arranged by others every day, the child's interest in the world may be worn away, and the spark of curiosity will be extinguished." Mi Qun recalled his own early experience of learning mathematics and programming at the Children's Palace, then studying physics at Fudan, then doing chip research and development at Intel, and later working at Google. "Can the specific knowledge I learned at the Children's Palace back then be directly applied to the frontier investment work I do now? Not really. But the curiosity and rapid learning ability that were protected and stimulated in childhood have benefited me for life." Mi Qun also advocates that children must be allowed to "play" more, explore more, and even experience failure more. "If they fall down or fail, parents should not be anxious, and even less should they protect them like flowers in a greenhouse. Experiencing setbacks and allowing failure is precisely helping children build resilience. Overprotection may instead cause children to lose the opportunity to face difficulties. Play is likewise not the opposite of learning." Mi Qun said that the human brain is very powerful; when children observe, try, and explore through play, the experience they gain does not require feeding massive amounts of data first, as in machine training. "After seeing a cat once, a child can form an understanding of cats in their mind. This kind of perception and exploration from real life is an indispensable part of growing up." He also pointed out that the future forms of work are full of uncertainty, and today's existing occupations may soon be replaced by new ones. "Rather than fixating on currently popular majors to plan a child's future, it is better to help children find what they are truly good at and love, and use curiosity to stimulate their learning ability and desire to explore. Only in this way can they do something to the utmost without feeling tired, and at that point opportunities will naturally emerge." Being smart is a gift; being kind is a choice. "Stay kind. Only if you stay kind will the AI of the future be kind." At the graduation ceremony, Mi Qun offered this message to the young people present. He mentioned that one of the core challenges facing the global AI community today is the "alignment" problem of large models. Simply put, it is ensuring that the behavior, goals, and values of artificial intelligence remain consistent with human expectations and interests. "AI is getting smarter and more capable, but how do we ensure it does not 'do evil' or 'go out of control'? When thousands upon thousands of agents begin to communicate and collaborate on their own, if their values have not been properly guided, the power of technology will cause harm." Mi Qun said bluntly that a large number of white-collar jobs relying on logical reasoning will most likely be replaced by AI in the future. "When computers long ago surpassed humans at chess, the genuine emotional connections and empathy between people have instead become the scarcest abilities. If a person is very self-centered and does not care at all about others, then no matter how smart they are, they will not go far on this path, let alone be happy." Wang Yanfeng also borrowed a famous quote from the founder of Amazon to remind the audience: "Smart is a gift, but kind is a choice. You must choose to be a kind person, so as not to let down the smart brain heaven gave you, and not to let down the cultivation of your country and your university."