What we're talking about now are data centers that are in the next several years going to use more power than the previous 50 years on Earth took. We're talking about individual buildings the size of football fields that have more power coming into them than midsize cities. And they're being built across the US. They're being built in Canada. They're being built throughout the Nordics. are being built here in Paris and throughout France, in Europe, in the Middle East in nations that sort of weren't front and center in anybody's mind previously. You know, Kazakhstan, Tajjikstan are building out Georgia building out data centers of size.

This paints a concrete picture of the unprecedented physical and energy scale of the current AI infrastructure buildout, grounding abstract hype in stark material reality.

The irony is unlike many sort of exciting times in technology. They're trying to capture yesterday's demand, right? The demand is way outstripping our ability to build data centers and to fill them with hardware. We have a $25 billion backlog, and we are not alone in that — OpenAI, Anthropic, Google wants more data centers, Microsoft wants more data centers, AWS wants more data centers. All of these players are not chasing sort of if you build it, they will come. They're chasing the demand is booked.

A $25 billion backlog signals that AI compute demand is not speculative but already contracted, which is a fundamentally different market dynamic than previous tech booms.

We have a term for token maxing. And there's a great debate. Is this actually creating value? I liken this to when we first started with AWS and it was so good to get around your own IT organization that you told every engineer, yeah, go ahead, put on your credit card and sign up. And a lot of it was really useful and some of it was like, God, I wish we didn't do that. But it doesn't mean that the net value isn't enormous. It means some of it is going to go nowhere.

The AWS analogy reframes current AI over-consumption not as waste but as a predictable early-adoption pattern that eventually matures into disciplined, high-value use.

We're really seeing a certain type of person emerge who knows how to deploy this technology. Systems thinking — which developers kind of have innately, CEOs tend to be great strategists and understand systems. But the intelligence is getting so much better every step along the way that I'm watching individuals typically startup founders but also venture capitalists and associates who work at my venture firm. They start playing with the tool and then the tool starts playing with them. They start to go, 'Oh, I haven't clearly defined what my goal is. I don't understand what a system is. I don't understand — I've never heard about making a requirements document.' And the software's like, 'Do you have a requirements document? What's your goal?' The AI starts telling people, 'You're token maxing and you need to get a little more focused here.'

This identifies a new professional archetype — the systems-thinking AI operator — and reveals how AI is forcing even non-technical users to develop rigorous problem-structuring skills.

At first, prompting was like that, right? You modified your prompt a little bit and it changed the answer dramatically. And increasingly it's understanding what your intent was. And if you have a chance to play with Fable or 56 from OpenAI, increasingly what you don't have to get the prompt just right. You don't have to be a prompt whisperer. Instead, you ask it and it says, 'Well, here are some things and by the way, maybe you wanted the chart to go two ways. You wanted a line and a bar.' And it's like, 'Well, that's exactly what I wanted. I didn't ask for it, but that is better.' And so it's understanding intent and that's a huge leap which if we were sitting here two years ago, the idea — we would never have been able to predict in a short 24 months that it would go from being a great summarizer researcher of web results to actually understanding your intent and then providing a solution and abstracting it all from you.

The shift from prompt-literal to intent-aware AI represents a qualitative leap in human-computer interaction that makes the technology accessible to billions of non-technical users.

Unlimited tokens, I believe, means unlimited reasoning. What does that mean? If you run these for 25 or 48 hours, you get amazing things now, and what if by using Cerebras we were 15 times faster and then you ran it for 24 hours, right? And you got weeks or months worth of thinking.

This reframes compute speed not just as efficiency but as a multiplier on the depth of reasoning available within a fixed time budget, with profound implications for scientific discovery.

AGI — I think I suspect you'll agree with me that we've hit it. We just haven't exactly deployed it fully. We have artificial general intelligence now. It feels like when we're talking about these reasoning moments and the ability for it to be as smart as any human — by any definition we had 20 years ago we've hit it. I mean, you think about that — any period of time sort of 10, 15, 20, 30, 40, 50 years ago, any definition we would have previously put forward — we've blown past it. And so which goes back to our previous point of like, do we know the questions to ask — 20 years ago, science fiction authors had their say and we answered all their questions.

A sitting AI CEO declaring AGI has already arrived challenges the field's framing of superintelligence as future-tense and forces a reckoning with what benchmarks actually matter.

Powerful recursive gains are exponential, right? You get better, you do it again. And if you continue to get gain, the slope of that curve is so steep. And that we're just beginning to see that now. You ask it a question, you learn from the results, you ask it to do it again, the results get better and more information is added. Your answer gets better. You ask it to do again, it covers more material. And these sort of loops are producing sort of not a little bit better answers but vastly better answers. And that is enormously powerful because we don't quite know where it ends.

The recursive self-improvement loop is the core mechanism behind fears of an intelligence explosion, and hearing a chip CEO describe witnessing it empirically makes the theoretical concrete.

The problem with human learning is it often moves at the pace of a generation. And like elephants and other large mammals, we don't have generations but every 15 or 20 years. And if you want to move really quickly across generations, you want them happening more like Drosophili, like fruitfly. You want two a day. Right. Then you see that in genetics, that's why we study them in genetics because learning encoded in the DNA, you can study over thousands of generations. And I think that what we're getting is that equivalent in AI. We're getting sort of learning so quickly over the equivalent of thousands of generations.

The Drosophila analogy elegantly captures why AI progress feels qualitatively different from human institutional learning — it's compressing generational timescales into months.

We have a shot with this technology so not our children nor anyone they know dies of cancer. There will be some dislocation in the economy. Sure. There was dislocation when cars came and it was a bad deal to be a guy who shed horses, right? Or built carriages. But you got to also against that, you know, make your tee of the cons and the pros. There's a shot that our children, none of them nor their people they love will die cancer. And that's one thing that we can work on with this technology. Unlimited energy, unlimited calories, unlimited knowledge, unlimited education, unlimited housing.

Grounding AI's upside in the elimination of cancer and scarcity reframes the public debate from job displacement fears to civilizational abundance potential.

We invented an algorithm called latent diffusion which is basically like the fundamental algorithm behind all of like generative models that are being deployed for image generation, video generation, even like physical AI now. It basically makes use of this principle that you can compress natural data such as images, such as video, such as audio into a much more like efficient representation and then train a transformer model on that.

Understanding that latent diffusion is the shared foundation beneath image, video, audio, and physical AI models reveals how a single algorithmic insight is powering an entire generation of AI products.

I think it's important and that's at least like the view that we have is that these AI models, they are a medium, right? We don't want to set like any way of how they are supposed to be used. We don't want to tell anyone especially not someone like Martin Scorsese how is he supposed to use these models. He is one of the greatest filmmakers ever. It was insane sitting in the same room with him multiple times and actually him seeing like exploring our models as one of the like core researchers behind it was like just an insane feeling. I think it was really this idea of like he has clearly a vision in his head of like a scene or a scenery where like maybe a new movie will be shot and he's trying to explore that — getting the mental picture of something out of your head and communicating it in a visual way by making these images is something that just makes it easier to communicate and convey like an idea of like what is actually in your head.

Framing AI image generation as a new communication medium rather than a replacement tool reorients the creative-AI debate around augmentation of human vision rather than automation of craft.

When I look at multimodal generative models as a whole, I think what really excites me is you can use the same kind of AI model to make a movie and deploy that as a brain on a robot. And I think this is so interesting. The technology is so powerful and so versatile — all the talk around world models, world action models — all of that, it's basically all the same and I think that's what's making it so interesting and what I find like most exciting.

The convergence of video generation and robotic action prediction into a single model architecture suggests the path to general-purpose physical AI may already be underway.

it is like it's becoming much faster. It's becoming more interactive. I can imagine like a whole bunch of like very interesting interactive content creation tools that you could host on Disney Plus Plus or elsewhere.

The shift toward real-time, interactive AI-generated content signals a fundamental change in how entertainment platforms could be structured.

I think the most interesting thing I've seen in this regard is uh fan films, right? So there's a category before generative AI fanfiction. People would write their own Star Wars story. Then there came fan films where people would dress up as Jedi Knights and record their own films.

Tracing the evolution from fan fiction to fan films frames AI-generated content as the next natural step in participatory fan culture.

George Lucas said, 'As long as you're not doing it commercially, you're not selling it, I give you permission to go make Jedi movies.' And they even released how to, you know, how-tos on how to make a lightsaber or, you know, sound files of like how to make the lightsaber sound.

George Lucas's permissive non-commercial fan-film policy offers a historical precedent for how IP owners might structure AI content licensing.

Now, people are taking the stories that haven't been told from the Star Wars universe and they're recreating them using AI. And for the fans, they're becoming quite popular on YouTube. Uh Star Wars Stories Untold is, I think, the biggest one. It's getting millions of views per video already.

Millions of views on AI-generated fan content validates real audience demand and signals that the market for IP-licensed AI storytelling is already forming.

I think that's really the future is letting the customer base pay a licensing fee or pay a fee, uh maybe rent software or maybe based on the output and let them be creative with the characters, let them make their own stories, and you could be in a unique position to empower that.

Proposing output-based or software-rental licensing models for AI fan content points toward genuinely novel IP monetization structures that don't yet exist.

if you find like a model that works for like the um IP owners uh but then also can enable like these super like creative customization use cases I think that's great. Yeah. I mean like like I mean like for myself like I when I read a book or whatever like watched the movie I had like so many like ideas how it could be done differently or this could have happened right this is like so nice that you can actually enable people to visualize these ideas.

The guest grounds the abstract IP licensing debate in a deeply human creative impulse — the desire every audience member has to reimagine the stories they love.

we raised a bunch of money. Uh we just crossed 100 people. We're hiring in Germany and in San Francisco.

A rapidly growing, dual-geography team signals Black Forest Labs is scaling internationally rather than concentrating solely in Silicon Valley, reflecting the global nature of AI talent.

Yeah. Um on the one hand, we're always looking for researchers who have experience in large scale model training. Um experience in diffusion model training, flow matching training. We're looking for engineers who want to be working with the customers to you know develop these like customized um physical AI solutions.

The hiring priorities of a frontier AI lab reveal which technical capabilities — flow matching, physical AI — are genuinely scarce and strategically critical right now.

We are looking for engineers who have experience in just like large scale compute infra managing that um and making sure that the training runs runs smoothly that we maximize our MFU and all that.

Mentioning MFU (Model FLOP Utilization) as a hiring criterion shows how infrastructure efficiency has become a first-class competitive concern for AI labs.

it seems like the corporates really want to have some additional level of control, but they also need the frontier models or your proprietary ones for some of those refined features. So, I think you have a very bright future.

This tension — enterprises wanting control but needing frontier capabilities — defines the core strategic opportunity for open-source-plus-proprietary hybrid AI companies.