Openai CEO Sam Altman is bullishing that not only will general artificial intelligence (AI exceeding human capabilities) be achieved in the very near future, but there may be breakthroughs that can only be achieved through the use of AI.
In a wide range of candid interviews covering everything from how Altman thinks his kids interact with technology to those without ChatGpt, Altman was probably surprisingly optimistic about his ability to align technology with human values.
He also gave a glimpse into Openai's hardware ambitions, hinting at the ultimate “family devices” and expressed other meditations about the future of AI.
Jan Philipp Burgard of Welt-in-Chief interviewed Openai CEO Sam Altman this week on behalf of the Axel Springer Global Reporters Network.
The following Q&A has been edited for length and clarity.
Welt: Sam, what quality can artificial intelligence be replaced?
Sam Altman: I think we know. But what's really unique is how much people care about others, what they do, and how much they want to interact with others.
I think these qualities will become increasingly important in the AI world. There are incredible tools, but you need to understand what to do, what others want, what others think will be useful.
Artificial intelligence is rapidly developing. Do you think there is always a clever tension than humans in all respects?
In many ways, GPT5 is already smarter than me, but I think so too are many others.
The GPT5 can do incredible things that many people find very impressive. But it also doesn't do much that humans can easily do.
That has been the case for some time, as humans use these tools and apply human insight, creativity and ingenuity in truly important ways. However, we expect the trajectory of AI capabilities will remain very steep.
Only three years after ChatGpt was released, the model has become much more capable. And I don't see any signs of slowing down. Over a few more years, it becomes very plausible for AI to make scientific discoveries that humans cannot make their own. For me, it's going to start to feel like something we can properly call super intelligence.
Is there an exact year to expect this super intelligence to emerge?
I would certainly be very surprised if by the end of the decade, by 2030, there was no very capable model to do what we ourselves cannot do.
Related Stories
Business Insider tells the innovative stories you want to know
Business Insider tells the innovative stories you want to know
I would also be surprised if I didn't see the same rate of progress seen in 2026 in 2024 and 2025. This means that by the end of 2026, models can be expected to be something very surprising, even today.
What percentage of work today could disappear in the near future?
Think about a job you did 30 years ago that might not exist at all today, or a new job that was hard to imagine 30 years ago.
I read statistics that about half of society's work changes every 75 years. That's even without AI. It may happen, and I hope it happens faster than it is now.
I think it's convenient to think about tasks rather than work percentages. There are many jobs that mean that the job will change. Of course, there's a whole new job. And many existing jobs will disappear completely and be replaced by these new jobs.
But what's even more interesting is that in everyone's work, what percentage of the tasks they do every day is done by AI. It is easy to imagine a world where 30-40% of the tasks taking place in today's economy will be carried out by AI in the not too distant future.
You became a father this year. What kind of education do you recommend your son to pursue his job so that he will not be replaced by AI in 30 years?
Metaskills of learning methods, learning to learn to adapt, learning to be resilient to many changes. I mentioned this before, but I learn how to understand what people want, how to create useful products and services for them, and how to interact with the world.
I am sure people are still at the heart of each other's stories. Anything in that world would be wonderful. We also believe that human desire for new things, their usefulness for others, and their desire to express our creativity is all infinite.
People rightly question what we are all trying to do in all these previous technological revolutions. These machines came during the industrial era. We saw them do what we were doing and said, “What is our role?”
Each new generation uses their creativity, new ideas, and all the tools that the previous generation has built to surprise us. I'm sure my kids will do the same thing too.
You sound very optimistic, but there are also AI critics. For example, researcher Eliezer Yudkowsky says that the relationship between superintelligence and humans is roughly the same as the relationship between humans and ants. Are you afraid that AI will ultimately consider us as ants and simply destroy us?
I've heard many people explain many different ways of imagining the relationship between AI and humanity.
My favorite was my co-founder, Ilya Satsukeiber. He once said that artificial general information wanted the way that humanity was handled, or that all Agis treated humanity, like a loving parent. When we ask that question, we are kind of personification Agi.
I think this tool is very capable. Even if it's not intentional, asking them to do something can result in results that we don't understand. Therefore, it is extremely important to align with human values. I don't think they treat humans like ants.
Related Stories
Business Insider tells the innovative stories you want to know
Business Insider tells the innovative stories you want to know
Critics accused Openai of turning it from a non-profit institution into a commercial enterprise, and partially ignoring security risks in the process. Do you agree with this criticism? Or do you think sometimes you have to move on to make progress?
First of all, we still have nonprofits, and we always do so. We hope we have the best resource of all time and the most impactful nonprofits that can be. This is extremely important to our mission.
The governance role of this is also important to our mission, to ensure that we stick to our mission and prioritize safety, happiness and the greatest interests of humanity.
I obviously made some mistakes because I understand that using this new technology will make more in the future. But overall, I am extremely proud of my team's track record by understanding how to distribute these services safely, widely, widely, and widely.
One of our core beliefs is to find a way to build this tool, align it with human values, put it in people's hands, and express everything they want to do.
The world is seeing Openai's hardware innovation. You hired an Apple Designer for this. You can assume that your device will look better, but what does that do?
That would be cool.
Two major revolutions have occurred: the mouse and the keyboard. This is an idea of a monitor to display this type of window system. Certainly it was a breakthrough. Next, I got my touch device. This made it a very personal device that lets you take out the mouse and use your fingers. This was huge.
Essentially, we have never had anything as powerful as AI. Computers can really understand what we want, which has allowed us to rethink the meaning of using a computer. So we're still exploring. It will take quite a while. Don't expect anything right away.
But over time, we expect to create a small family of devices. Certainly they will look good, but that's not the main thing. If we really do a great job, they hope that it changes the meaning of using a computer, the way you work, and how you play and live your life. But there are a lot of work and lots of work everywhere.
What is one feature or example of a feature that such a device can have?
Currently, if you want to perform a task on your computer, you will need to click and navigate between many applications. If it's a difficult task, it may take a while. One of the promises of AI is that we can say complicated things that need to happen in a day, a month, or even a year.
You can trust that the computer will understand it, do it for you and come back to you when you need help. But you can imagine asking a computer a very subtle and complicated but simple question and trusting it to do the right thing and come back to you when you need help. It would completely change what it feels like to use a computer instead of launching lots of apps and constantly appearing notifications. That's one thing.
The US tech industry has long had a liberal democratic tone. Many tech leaders are now showing support for Donald Trump. How would you explain this change in atmosphere?
The tech industry needs to work together no matter who the US president is. But I think there was a welcome policy change in this particular case. The ability to build infrastructure in the US is extremely difficult and extremely important for companies like ours. President Trump has done a great job of supporting this. In general, more professional business and protect climates were also a welcome change.
The United States is very biased and indeed divided. What do you think about the idea of having AI govern on behalf of the US president in the future?
I don't think people want that anytime soon. But what I'm hoping for is that presidents and leaders around the world increasingly use AI to help them with complex decisions. But I think we all still want humanity.
To finish off, many people get relationship advice from ChatGpt. Have you ever asked your bot to help with a relationship issue?
I don't use it as much as other people do. I've tried it, but no, it's not one of my big personal use cases. But that is clearly something that many people use.
