Trường học thời A.I. sẽ ra sao? | TS. Phạm Hy Hiếu, CEO One Nexus, ex- Open AI, xAI, Google Brain

Summary

The episode explores how education, parenting, and schools may change as AI becomes more capable. Guest Pham Hy Hieu draws on his experiences in school and technology companies. Many of his recommendations are personal reflections rather than universal rules.

1. Motivation can grow from small experiences

• Hieu says his interest in mathematics developed gradually. As a child, he was sent to extra classes after struggling with a test. There, a teacher gave him problems he could solve, and quick, competitive classroom exercises made math enjoyable.
• He distinguishes between wanting to win and hating to fail. He describes himself as more afraid of failure than attached to winning, and says that fear pushed him to work on weak areas.
• Hieu and host Trang discuss the risks of turning motivation into excessive pressure. His example of Elon Musk raises a question for parents: exceptional drive may come with significant personal costs, and parents should not assume they want their children to follow the same path.
• Trang’s parenting approach is to give children resources and opportunities without deciding in advance what they must become. Hieu raises a contrasting possibility: parents may sometimes guide children toward a focused pursuit. The discussion does not settle on one answer, but emphasizes that different children and families may need different approaches.

2. Gaming: Set boundaries, but recognize its potential

• Hieu recalls being deeply absorbed in video games as a child. He says he still avoids returning to one game because he fears losing too much time to it.
• He suggests that parents set limits rather than impose a total ban. One example is allowing a set amount of daily play while encouraging a child to learn programming or use coding tools. He jokingly proposes making the limit something a persistent child could try to overcome by learning how the restriction works.
• In his view, gaming becomes a problem when it interferes with other commitments—for example, when a child neglects assignments, lies, or gives up other activities. He also cautions that excessive gaming during childhood can displace exercise, social activity, and healthy development.
• Games can also offer enjoyment and a sense of challenge. Hieu sees a possible connection between gaming and technology work: both may appeal to people who like competition, experimentation, and finding ways to achieve a goal. He presents this as an observation, not proof that gaming causes success in technology.
• The conversation also notes that game design deliberately uses techniques to keep players engaged, making it unsurprising that some people find games hard to put down.

3. Learning momentum matters

• Hieu compares learning to running a marathon: the first stretch is often the hardest, when the subject is unfamiliar and progress feels slow. Once learners have built enough knowledge to understand the field and engage with it, learning can become more rewarding.
• He therefore values continuity and momentum. In retrospect, he says he might have preferred to continue studying rather than take a year off to prepare applications to U.S. universities.
• Trang points out that Stanford also gave him exposure to people, technology, and career possibilities he might not have encountered otherwise. Hieu acknowledges those benefits, but feels that the personal cost and loss of learning momentum outweighed the return for him.
• Their discussion suggests there is no universal answer: a school’s opportunities can matter, but outcomes also depend on the individual, their peers, and what they do with the experience.

4. School prestige and degrees are not the same as ability

• Hieu says that, in his experience at leading technology companies, a school’s name is less important than what a candidate can actually do.
• He describes interviewers asking applicants about a project they are proud of, then probing several levels deeper: what they built, why they made particular decisions, and how the work functions.
• A strong candidate should be able to explain their contribution and show evidence of it, such as code or other project outputs. The interview is meant to reveal whether the person understands the work and can perform the job.
• Hieu says degrees and grades may still influence hiring in Vietnam, but believes project-based evaluation will become increasingly important as AI automates more routine work.
• He also resists defining himself by a university or company brand. His view is that a person’s identity should rest on their own actions and abilities, not on the institution they attended.

5. AI could make learning more personalized

• Hieu argues that education should teach students how to use AI responsibly rather than prohibit it outright, since AI is likely to be part of work and daily life.
• AI could support highly personalized learning by adapting explanations and teaching methods to a learner’s interests and ways of understanding.
• He illustrates this with his own view that conventional music notation feels inefficient to him. AI, he suggests, could help teach music theory in a format that better suits a particular learner, just as it could adapt explanations in other subjects.
• Such personalized support is currently more accessible to families who can afford private tutors. Hieu sees AI as a way to make tailored instruction available to more learners.
• The conversation leaves questions of academic honesty and fairness open, but stresses that schools need policies and teaching practices for AI use rather than assuming it can simply be excluded.

6. Schools may shift from delivering knowledge to supporting social development

• Hieu questions whether schools’ traditional roles—teaching content and granting credentials—will remain as central when AI can teach and assess learners.
• The COVID-era experience with remote learning is discussed as a warning: online lessons may transmit information, but they cannot fully replace direct interaction with teachers and peers.
• Hieu expects schools to retain value as safe, social environments where children meet, play, build relationships, and practice communicating with others.
• Trang describes school as a kind of small-scale society that can help prepare students to participate in the wider one.

7. People still need to learn in an AI era

• Hieu challenges the assumption that intelligence alone should determine who is in charge. If AI becomes more capable than humans, he argues, people must still ask where human value lies.
• Learning will be important for interacting and collaborating with AI, contributing to the development of knowledge, and taking part in discussions about human purpose and values.
• The conversation also considers philosophical and religious reflection. Hieu says AI can produce thoughtful language, but its output does not replace a person’s own experience, reflection, or search for meaning.
• His broader point is that learning remains necessary not only for employment, but also for understanding ourselves and deciding what matters.

8. Hieu’s practical learning method: Learn with a purpose

Hieu identifies one core method: study, work, and live with a clear purpose.

• At a given time, identify the most important goal you are trying to achieve.
• Prioritize the learning and tasks that directly support that goal.
• Be willing to make trade-offs. Hieu gives the examples of focusing on improving at geometry or prioritizing relevant coursework when pursuing an internship.
• Reduce time spent on tasks that do not serve the current goal, while recognizing that rest and recreation still have a place.
• Revisit the goal as circumstances change. The purpose is to use time deliberately, not to pursue one fixed target forever.

Speakers and sources featured

Featured speakers

• Lê Diệp Kiều Trang — host and interviewer.
• Dr. Phạm Hy Hiếu — guest; AI researcher and technology executive.

People, organizations, and examples mentioned

• Elon Musk — discussed as an example of exceptional work motivation and high-pressure leadership.
• Hieu’s parents, especially his mother and father, and his mathematics teacher Nguyễn Hữu Vĩ — mentioned in accounts of his childhood and learning.
• László Polgár and his daughters Sofia, Susan, and Judit Polgár — discussed as an example of intensive parental guidance toward chess.
• Magnus Carlsen — mentioned in connection with Judit Polgár.
• AI tools and companies — ChatGPT, Claude Code, Google Brain, OpenAI, xAI, Anthropic, Tesla, and Meta.
• Other referenced settings and examples — Stanford, the National University of Singapore (NUS), COVID-era remote schooling, and a Catholic priest’s use of AI as a prompt for reflection on personal meaning.