Sam Altman on the path to AGI
Altman sees AGI as a poorly defined mile marker on one long exponential, not a finish line, with the real milestones being systems that discover new science and out-research human experts. He expects steady progress from scaling, limited by compute, energy and the slow pace of real-world experiments. He argues for short timelines with a slow takeoff, made safe by deploying iteratively, and for people to build adaptable skills.
15 points · 7 appearances · 2023–2026
How the view evolved
AGI is a poorly defined mile marker → The exponential goes past AGI
In 2024 he called AGI a poorly formed question and a mile marker closer to a beginning; by TED 2025 he argued the conversation should move off the AGI moment entirely toward building a society and safety approach for a system that keeps going far beyond it.
Up next
2023–2026 · changed view
“I love that you asked them together, because I think they're super related. I think the economic transformation will drive much of the political transformation here, not the other way around. My working model for the last five years has been that the two dominant changes will be that the cost of intelligence and the cost of energy are going, over the next couple of decades, to dramatically, dramatically fall from where they are today. And the impact of that, and you're already seeing it with the way you now have like, you know, programming ability beyond what you had as an individual before, is society gets much, much richer, much wealthier, in ways that are probably hard to imagine. I think every time that's happened before, it has been that economic impact has had positive political impact as well. And I think it does go the other way too, like the socio-political values of the Enlightenment enabled the long-running technological revolution and scientific discovery process we've had for the past centuries. But I think we're just going to see more. I'm sure the shape will change, but I think it's just long and beautiful exponential curve.”
2023–2026 · changed view
“So I want to be very clear. I do not think we have yet discovered a way to align a super powerful system. We have something that works for our current scale called RLHF, and we can talk a lot about the benefits of that and the utility it provides. It's not just an alignment, maybe it's not even mostly an alignment capability. It helps make a better system, a more usable system. And this is actually something that I don't think people outside the field understand enough. It's easy to talk about alignment and capability as orthogonal vectors. They're very close. Better alignment techniques lead to better capabilities, and vice versa. There's cases that are different, and they're important cases, but on the whole, I think things that you could say like RLHF or interpretability that sound like alignment issues also help you make much more capable models, and the division is just much fuzzier than people think. And so in some sense, the work we do to make GPT-4 safer and more aligned looks very similar to all the other work we do of solving the research and engineering problems associated with creating useful and powerful models.”
2023–2026 · changed view
“People talk about like how many jobs is AI gonna do in five years and the framework that people have is what percentage of current jobs are just gonna be totally replaced by some AI doing the job? The way I think about it is not what percent of jobs AI will do, but what percent of tasks will AI do and over what time horizon. So if you think of all of the like-five second tasks in the economy, five-minute tasks, the five-hour tasks, maybe even the five-day tasks, how many of those can AI do? And I think that's a way more interesting, impactful, important question than how many jobs AI can do because it is a tool that will work at increasing levels of sophistication and over longer and longer time horizons for more and more tasks and let people operate at a higher level of abstraction. So maybe people are way more efficient at the job they do. And at some point, that's not just a quantitative change, but it's a qualitative one too about the kinds of problems you can keep in your head. I think that for videos on YouTube, it'll be the same. Many videos, maybe most of them, will use AI tools in the production, but they'll still be fundamentally driven by a person thinking about it, putting it together, doing parts of it, sort of directing it and running it.”
2023–2026
“When I was a kid, I assumed that there were always some adults in the room. Someone had a plan. Someone knew everything that was going to happen. Someone had it all figured out. And I sort of think why people like conspiracy theories is it's nice to think that someone's got a plan. It's nice to think someone that, you know, has it all figured out. And then I got a little bit older and I sort of started to suspect there are no adults in the room. No one, people have plans. I have plans, but no one has all the answers. No one knows where it's all going to go. And now that I am the adult in the room, I can say with certainty, no one knows where it's all going to go. Like, I'm the guy in the room and I have some guesses and I have some plans, and we're working really hard. But like, you know, we try to always say what we think the possibilities are, what we think is most likely. Often we're right. Sometimes it's in the broader set. And sometimes it goes in a totally different direction than anything we thought. And you know, we keep trying to make progress, figure out more. We try to tell people, not just tell, we try to show people by like deploying these systems and say, "You can go use it. Don't just take our word for it. Try it out. See what it can do."”