Ultraman on AI entrepreneurship: Where are the opportunities for young people and the next big thing?

Key points:

GPT-5 has achieved "integrated intelligent services" and can replace traditional workflows that require professional teams. In the future, entrepreneurs can build high-value companies with just a few people plus AI tools.

Currently, the most important thing is to learn to use AI tools proficiently. The ability to learn quickly is more crucial than professional skills. Practitioners should focus on solving real needs rather than blindly following technological trends.

Even if AI becomes smarter than humans in the future, the emotional connection between real people will become even more precious. Humans make mistakes, have unique experiences, and possess rebellious thinking, which AI can never imitate.

At the early stage of all technological revolutions, there will be a large number of "wrapper applications," but only those companies that truly solve users' pain points can survive. Just as Uber has profoundly changed the way people travel, it is far more valuable than a simple flashlight app.

Altman believes that robots will become increasingly important and are best designed in a humanoid form because our world is originally designed for humans. Future AI assistants will require new hardware forms.

AI can indeed help startups quickly complete tasks such as development, customer service, and marketing, but relying solely on AI cannot build a defensible business. The key is to build real and valuable services around AI.

If you're in your 20s today and still at the starting point of your career, which path would you choose? Altman gave a straightforward answer: This is the most exciting era. With the support of a doctoral-level AI assistant like GPT-5, one person can accomplish what used to require decades of experience and a large team. Whether it's starting a business, writing code, or exploring scientific research, you can achieve far more than any previous generation of young people.

Recently, Sam Altman, the CEO of OpenAI, had an in-depth conversation with Nikhil Kamath, the co-founder of India's largest brokerage firm, Zerodha.

In this conversation, Altman talked about how GPT-5 is reshaping the paths of learning and entrepreneurship, and how young people can use AI to achieve what used to take decades. He also emphasized the truth of entrepreneurship - there is no perfect plan in the world, and everyone is "learning by doing."

The following are the highlights of Altman's latest interview:

01 GPT-5 Becomes an All-Round Doctoral-Level AI Assistant

Question: OpenAI recently released GPT-5. What's different about it?

Altman: It's hard to simply describe how good it is, or in which aspects it's better than GPT-4, or the programming demos it shows that GPT-4 can't do.

I've gotten used to the fluency and adaptability of GPT-5. It has a comprehensive intelligence that none of our previous models had. It's an integrated model. You don't have to think about whether to choose GPT-4, GPT-3, or GPT-4-mini. It's a model that's ready to work at any time.

For example, it's like an assistant that provides doctoral-level expert services at any time. You can ask it questions or let it help you with any task. For instance, if you need to develop a software, it can help you start from scratch; if you need a research report on a complex topic, it can handle that too; if you need to plan an event, it can also get it done.

Question: In essence, is this function more autonomous and able to process tasks sequentially, thus getting you closer to your goal?

Altman: GPT-5 has become more "intelligent" and reliable in executing multi-step, long-term tasks, which is very helpful for workflows that require complex processes. Its performance in terms of persistence and complexity is truly impressive.

Question: Is there anything specific you want to talk about regarding GPT-5?

Altman: GPT-5 really represents a significant advancement for us. It has made remarkable improvements in terms of capabilities, robustness, and reliability. Whether it's developing software, answering questions, learning, or improving work efficiency, using it to handle these tasks is a very important step forward. Every time we see such technological progress, we're deeply amazed by the human potential it unlocks. Especially in India, which has now become our second-largest global market and may soon become the largest.

We've received a lot of feedback from Indian users about the services they want us to improve - including better language support and more cost-effective access methods. We've incorporated this feedback into the model's upgrade and optimized ChatGPT. We'll continue to work hard in this area in the future.

Every time we enhance our capabilities and provide users with more powerful functions, we're surprised by how young people use GPT-5 to start new companies, improve their learning methods, and get better medical advice.

02 Creativity + AI = Achieve in One Person What Used to Take Decades

Question: If I'm a 25-year-old young person, what should I do now? What should I learn? If I want to start a business, what kind of company should I start? If I'm looking for a job, which industries do you think will have growth potential in the next three to five years?

Altman: I think this might be the most exciting time to start a career. For young people, they can do more today than at any time in history. It's really incredible. With the current tools, you can achieve far more than what was possible for me when I was young.

Back then, there was also a similar enthusiasm, but the available tools were far less powerful than they are today. We've gone through the computer revolution, and what a 25-year-old could do was already unprecedented. Now, the rise of artificial intelligence is driving this change at an even more astonishing speed.

Whether you want to start a business, become a programmer, or enter other industries like new media, you can now use your creativity and ideas to accomplish tasks that might have required decades of experience or a large team in the past. This is an unprecedented opportunity.

In terms of industry direction, I'm particularly optimistic about the application of artificial intelligence in the scientific field. In the future, personal discoveries and breakthroughs in scientific research will be greatly accelerated. At the same time, the way of programming is undergoing a huge transformation. We might even create new forms of software and entirely new skill systems.

For entrepreneurs, if you have novel business ideas, you only need a small team to have the opportunity to achieve what used to require a large amount of resources. This will make entrepreneurship more feasible than ever before.

Overall, it's like a brand-new canvas. Many past limitations have been broken. The only boundary is the quality of your ideas and the height of your creativity. And these advanced tools you have are the best assistants to help you turn those ideas into reality.

Question: Suppose I'm majoring in engineering, business, or art. Are there any specific fields I can study to develop scientific projects using AI?

Altman: I think the most crucial specific skill is learning how to use new AI tools. Learning itself is extremely valuable, and the ability to "learn how to learn" is a skill that will benefit you throughout your life. Whether you're majoring in engineering, computer science, biology, or other fields, if you have the ability to learn quickly, you can quickly adapt and enter new fields. And right now, being proficient in using AI tools, especially being able to use them smoothly and efficiently, has become extremely important.

When I was in college or even high school, I thought the most valuable skill to master was programming. At that time, I wasn't entirely sure how it would affect my future work, but it turned out to be a high-value ability. Today, learning how to use AI tools might be even more important than programming was back then. In the future, there will be a huge gap between those who can skillfully use AI to complete various tasks and those who can't.

Besides, in this rapidly changing world, adapting to change and maintaining resilience are equally important. Also, figuring out what people really need and finding corresponding solutions is the key to success. I used to be a venture capitalist, and many people would ask me: What's the most important thing in entrepreneurship? My former boss, Paul Graham, the founder of Y Combinator, gave an answer that has always stuck with me and has become the motto of YC - "Build something people want." It sounds simple, but many people have spent a great deal of effort and still haven't been able to achieve it. Those who truly achieve it often achieve great success.

Question: When you mention learning how to adapt and change and quickly master AI tools, is there a shortcut?

Altman: Yes, a very effective way is to use GPT-5 to quickly create small software. GPT-5 performs far better in this regard than any model I've used before. In the past few weeks, I've often been amazed at how it can help me solve various small problems in my life.

My method is simple: I'll let it generate an initial version and then start using it right away. Then, I'll give continuous feedback - like telling it "It would be better if this function could be like this" or "If you add this feature, I can do something else." Through this continuous iteration, my tool gradually becomes more in line with actual needs. This process not only helps me learn to use AI tools more efficiently but also allows me to understand their potential more deeply.

03 There's No Perfect Plan in Entrepreneurship, and the World Moves Forward by Learning While Doing

Question: Graham mentioned five founders worth paying attention to in a report in 2009. You were only 19 years old at that time, but he put you in the same category as Steve Jobs, Larry Page, and Sergey Brin. What did he see in you? What do you think made you stand out from your peers?

Altman: Indeed, there's often an element of luck in entrepreneurship, but we also worked very hard at that time. I think what made us stand out might be our consistent focus on the long-term vision and our ability to think independently. For four and a half years before launching our first product, we spent almost all our time researching how to make AI truly work, and our thinking was different from most people's. This spirit of independent perseverance without external feedback and our firm belief in our ideas were very important.

Question: I'm not sure if you've studied Machiavelli. He had an interesting view: A prince should appear pious, kind, trustworthy, humane, and honest, but doesn't necessarily have to be so. I've noticed in many of your interviews that you often emphasize that you're "not that great." Is this deliberately displayed humble attitude still applicable in today's world or the new world you're about to face?

Altman: I'm not sure if a humble image is suitable for this world, but I can share some experiences. When I was 19, I thought the adults who managed big companies were really smart and had everything under control. But now, standing in this position, I realize that almost no one really has a perfect plan. Everyone is learning by doing. No one can predict everything. Everything is moving forward in a changing environment, and many decisions are made through continuous learning from data and market reactions. This is the process of entrepreneurship.

During the entrepreneurship process, you'll have many goals you want to achieve, but things often don't go as planned. You might bet on the wrong people or the right ones; some technologies make breakthroughs, while others stagnate. You can only work hard and move forward step by step.

We've tried many tactics. Some have eventually evolved into strategies, while others have proven to be futile. The market will give feedback, and competitors will react. We have to adjust accordingly. Now I deeply understand that everyone is groping in practice and learning at work. This isn't just false modesty; it's the way the world works.

Question: Is such an image really suitable to show to the world? Does a humble attitude still work today?

Altman: I'm not sure if a humble image is suitable for this world, but I can share my experience. When I was 19, I thought the adults who managed big companies were very smart and seemed to have everything under control. But now, standing in this position, I realize that almost no one has a perfect plan. Everyone is learning by doing. No one can predict everything. Entrepreneurship is about continuous experimentation and moving forward through feedback and change.

During the entrepreneurship process, you'll set many goals, but things often don't go exactly as planned. Sometimes you bet on the right people, and sometimes you bet on the wrong ones; some technologies make rapid breakthroughs, while others stagnate. The only thing you can do is work hard and keep moving forward. We've tried countless tactics. Some have eventually evolved into strategies, while others have proven to be futile. The market and competitors will give feedback, and we have to adjust accordingly. In the end, I've realized that everyone is groping in practice and learning at work. This isn't false modesty; it's the real way the world works.

04 Post-AGI Era: The Return of the Value of Family and Community

Question: What's your view on the future of marriage, religion, and children?

Altman: I hope in a post-AGI world, the meaning of family and community will become even more important. No matter how you call it, I think the weakening of these basic social units is a real social problem, which makes me a bit uneasy. I'm not entirely sure why, but I hope we can reverse this trend. As people have more wealth, time, resources, and potential, family and community are obviously the two major sources of our happiness. I hope we can return to this direction and regain these basic values.

Question: If human desires, needs, and competitive games (in any aspect) are almost limitless, do you think that even if everyone has enough, we'll still want more?

Altman: I really think human needs, desires, and the pursuit of status games seem to be endless. This isn't entirely a bad thing. Maybe it's just a part of human nature. I think even if we have more resources, we'll still find new goals to pursue, and new competitive points will emerge. This is the driving force for human progress and exploration.

Question: Do you think the world will continue with the existing capitalist and democratic models? Suppose a company, like OpenAI, reaches 50% of the world's GDP. Will society allow this to happen?

Altman: I don't think this will happen. If it does, society might say, "We don't accept this situation and need to find a solution." I like to compare AI to transistors. Once, it seemed that transistors would occupy a huge amount of value, but in the end, they became the foundation for countless products and services. We use transistors every day without even noticing their existence. Many companies have created countless innovative products and profits through transistors, and this model is highly decentralized. So, I guess this might be the situation in the future, rather than one company occupying half of the global GDP.

Question: If a company really reaches such a scale, do you think it will be nationalized?

Altman: I'm not sure if a company will be nationalized, but I think as society becomes more prosperous and the technological landscape changes, measures such as wealth redistribution and social support in society will increase over time. I can't predict exactly what form it will take, but I guess there will be various social experiments in the future, such as new sovereign wealth funds, universal basic income (UBI) programs, and the redistribution of AI computing resources. These experiments might take different forms in different countries, but