The Ledger
Every statement that passed quotation and attribution checks. Mix any filter with any other: certainty 1/5, debate potential 5/5, or both at once.
why aren't all 846 resolved? a statement only gets an assessment when the public record can support or contradict it. opinions and what-ifs never can, and 0 checkable ones are still open, waiting for their date. predictions held up or didn't; assertions are supported or contradicted. on every card: ▮▮▮▮▮ certainty · ▮▮▮▮▮ debate potential. speakers are clickable
Schmidhuber: No physical laws prevent uploading human minds into computers
“There is no physical reason to reject the notion that this might be possible.”
Bosworth: The era of monolithic AI models ended around Llama 3
“The era of the monolithic model kind of died around Lama III launch. Like, the idea, like, here's one model, and just, like, let's just test how smart this model is, and that was how good it's going to be at lots of things. We're now in a world where when you'…”
Brockman: The ideal AI product will have almost no interface
“And I think that the question of what's the interface you want, what is the product that you want, is what we spend a lot of time thinking about. And the answer is you want almost no interface. You want no product. Right. You want this to be like, what's the i…”
Brockman: Model scaling anomalies always stem from implementation bugs, not limits
“Every time we've kind of run into a like, oh, this isn't quite scaling the way we expect, it's we have a problem. We have a bug that our math wasn't quite right, that, oh, our implementation Isn't quite matching the math, whatever the thing is.”
Banning security training for American LLMs will increase software vulnerabilities
“If our American LLMs know nothing about security, they will create more security flaws. So we cannot create a standard where American LLMs are dumb about security. That would be a humongous own goal.”
Stamos: Banning bug-finding in US models degrades code security
“We cannot set the standard that US AI models can't find bugs. That is a terrible, terrible, terrible standard. If you have a, if you are writing software with an LLM, it has to be able to understand what a bug looks like so it does not write those bugs.”
Brockman: The ultimate AI product interface is zero interface
“You want almost no interface. You want no product, right? You want this to be like, what's the interface between you and me, right? Just being able to talk to a persistent entity of some form that's able to go and accomplish goals for you.”
Product Experience Drives Enterprise AI Retention Over Raw Model Quality
“Better models don't necessarily create the switching for employees and users. It's about the product experience around those models.”
Hinton: AI in closed systems like math can improve without external data
“I believe, for example, in areas like mathematics, because it's a closed system you don't need data. You can just make conjectures and see if you can prove them and keep on like that. In that sense, it's a bit like AlphaGo where you can play against yourself.”
Hinton: Defining AGI as Equal Human Performance Across All Tasks Is Flawed
“So the whole concept of AGI that it's going to be equal to people at everything all at the same time doesn't really make sense to me. It's going to be better at some things, worse at other things.”
Hinton: Predicting next tokens and frames is sufficient to create smart systems
“The answer to question one is yes, if you can figure out how to change each connection strength, you can make systems that are very smart just by training on data to predict the next word, or to predict the next frame of a video, or to predict something about …”
Hinton: AI Regulation Is a Steering Wheel, Not a Brake
“Progress is like the accelerator, but regulation is the steering wheel. We want this stuff to go in the right direction, not the wrong direction. What the big AI companies are saying is let us develop this very fast car without a steering wheel. That's not a g…”
Costolo: Advertising is undefeated because ARPU far exceeds premium subscriptions
“Advertising is undefeated as a business model. What you end up finding is that the average revenue per user is just much, much higher than you can get from a premium subscription.”
Roy: AI economics are more akin to industrials than software
“The economics of a general or an AI company are just fundamentally different than software. That like, it's more akin to industrial companies or something, because you're paying for The compute costs. So the resources in can scale linearly or somewhat linearly…”
Dolen: Tech industry value proposition has shifted from software back to hardware
“You had a significant shift in the value proposition. Where was the value coming from? It went from software back to hardware.”
Dolen: AI shrinks SaaS moats by enabling DIY CRM and ERP
“AI shrunk those moats, and so you can now do CRM on your own. You can now do some components of ERP on your own. You can now do customer attraction on your own and marketing and that type of functionality. Well, that's a compressing feature, right? So in addit…”
Shevelenko: AI pricing will resemble Costco with memberships and compute credits
“AI is going to become a lot like Costco where you pay for the membership, right? And that gets you in the store. And that's actually the part of Costco's business that is, you know, the highest margin. And then you have, you know, everything you're buying in t…”
Brockman: Workers will become overseers managing fleets of AI agents
“The place we're going is one where you as A person doing work that you are the overseer. You are the CEO of almost this autonomous corporation, or, you know, of this fleet of agents perhaps is more, is, is the way to say it, and that they are operating accordi…”
Brockman: Open-source distillation cannot completely replicate frontier AI capabilities
“Now, it is also the case that it's not as simple as you can take the output to these models and distill and you have exactly the model of the same capability, It's just smaller and can run fast. If that were the case, we would just do that”
Véliz: Predictive AI creates self-fulfilling prophecies with zero error signals
“Self-fulfilling prophecies are like the perfect crime. Because it's like a murder weapon that disappears upon striking. It leaves no record. It creates no error signals. We will never know how that person would have fared, because they will never get the job, …”
Véliz: Algorithmic bias audits are limited by a lack of counterfactual data
“When you say, well, let's investigate for bias or investigate for inaccuracy, there is a limit to what we can do, because we will never have the counterfactual. This is not a randomized control trial, right? And you still have the problem that without clear cr…”
Véliz: Opaque algorithmic systems induce alienation and magical thinking
“We are building systems that are very Kafkaesque, that are impossible to navigate, and I don't know if you've had this experience in which they are becoming so alienating and so Kafkaesque That people start having, like, magical thinking about the algorithm, a…”
Michael: AI enables drone threat discrimination far beyond human capacity
“You could do much more, be more precise, be more specific about what you're going after, what you're defending how you know, and the precision is really what's interesting to me because if you can use AI to detect and discriminate and, by discriminate, I mean,…”
Levie: Automating video editing with AI is harder than coding due to evals
“Editing a video is like, you know, going to be actually in many cases, a harder task than coding. Because the, because again, the code right now is like, it has this great property of in the eval process, in the training process rather, you can instantly evalu…”
Levie: Any AI Lab's Lead on Breakthrough Models Is Only 6 to 12 Months
“Unless there's some so kind of closed proprietary research event and breakthrough that happens that just simply nobody else knows about, and we have no evidence that we've ever had one of those in AI, like, you know, these things just eventually sort of emerge…”
Bose: Customizing model weights is a waste of R&D resources
“Our sort of maximalist thinking is that the frontier labs are going to keep innovating in the level of reasoning and capabilities of their models, and so trying to create these customizations or adding our own token weights Is not a good idea and is a waste of…”
Bose: Forking open-source models leaves software companies 3–6 months behind
“If you create a fork and you're always three to six months behind what your, Competition could be doing. You know, like there'll be other companies, I'm sure, who are thinking about some of the challenges that we are addressing at Asana. I don't want to be thr…”
Hoffman: Tech companies cannot secure data center financing without insurance policies
“You cannot get financing to build these data centers without an insurance policy.”
Brockman: AI Progress Is Now About Execution Harnesses Over Raw Models
“And we talked about it a little bit in the case of the underlying models, but the thing that's really changed over the past couple of years has been that it's no longer just about the model. It's about the harness. It's about how does the model get context? Ho…”
Brockman: Pre-Training Capability Multiplies Through the Entire AI Model Pipeline
“Every single step of the model production pipeline multiplies. And so you want to improve all of them. And the thing that we see is we prove the pre-training. It makes all the other steps much easier. And it makes sense because it's a model is able to learn fa…”
Brockman: Massive Scale AI Training Requires Concentrated Compute Clusters
“Well, because the, there's multiple reasons but one is that even as the balance of how much inference versus training changes, that you cannot get massive scale training Through any other way besides this concentration of compute on one problem.”
Brockman: Treat AI Compute as a Revenue Center Like Hiring Salespeople
“You can think of compute not as a cost center, but as a revenue center. Think of it a little bit like hiring salespeople. How many salespeople do you want to hire? As long as you can sell your product, as long as you have a scalable way to sell that product, t…”
Yen: Gmail exists primarily to permanently log users into Google's tracker
“And why did Google go so quickly into email? The reason is because they figured out in order to get all your information and correlate it to a single profile, they need you to be logged in. And the one thing that you're always logged into all the time online i…”
Sorkin: Private equity and credit both delay markdowns to preserve valuations
“There is a whole sort of universe that I'd put in the category of mark to make believe. And the incentive system is such that you want that mark to make believe to go on as long as possible, because if you are The equity owner, you don't want to mark your mark…”
Sorkin: Post-WWII middle-class American dream was a historical aberration
“I actually think that was a historical aberration. That was not the way it used to be. If you look at the 19 twenties, it was not like that. You look at the 19 thirties, it wasn't like that. The leave it to be for American dream that I think this individual is…”
Foundation AI labs face constraints, creating massive gaps for independent startups
“These labs have so many resources, but they are still constrained. They're constrained on, like, compute. They're constrained on inference. They're constrained on people. Every second building, like, a new creative model is a second they could have spent on a …”
Fast-moving AI product teams beat domain experts in enterprise software markets
“I think in general though, like pre AI for if you were building, especially an enterprise business, say software for HVAC, like the people you would want to back in that market is the guy who has done HVAC You know, knows the market inside out, built a company…”
Pollan: Historical compromise left modern science ill-equipped to study subjectivity
“And now we'll take everything measurable and quantifiable. And the church, you can have everything subjective and qualitative. That was Galileo's deal and Descartes to some extent. And it was a very pragmatic deal, but it's left us with a science that's ill-eq…”
Adler: Recursive AI Improvement Requires Automating Research, Not Just Coding
“That's one step, but then you need to take that engineering and use it to actually automate the AI research. You need to go from being able to implement the ideas more quickly to using that to fuel faster and faster growth in the breakthrough ideas themselves …”
Adler: AI alignment cannot reliably encode human values into systems
“And the fundamental problem that we're seeing is we don't know how to take our values or our goals and encode them into these AI systems and get them to pursue it reliably.”
Traditional SaaS is just a database UI that AI will replace
“So at that point, you just have so much of traditional software Can just be looked at as like a UI over a database. That's what's happening right now, and the market is finally actually coming to that realization.”
Pineau: Reasoning models fail at multi-tier hierarchical planning
“That's the part that the reasoning models don't do. They do really well at like one level of granularity... But the going back and forth between different levels of sort of resolution of action, it's really hard. So on the technical terms, we call it hierarchi…”
Siegler: Nvidia was likely acting as OpenAI's infrastructure debt guarantor
“To me, that seemed like what a big part of this deal was basically NVIDIA stepping in to be a, you know, a guarantor of the debt that, that opening AI would need to raise.”
Siegler: Microsoft and Amazon back multiple AI labs to stop Google
“I think these other major cloud players, at least the ones who have the rival clouds, specifically Microsoft and Amazon realize that yeah, they probably need to align around basically anyone who's not Google, right? They don't necessarily care if it, I mean, t…”
Bret Taylor: Business Model Transitions Are Harder Than Technology Transitions
“I think the interesting maybe counterintuitive point that I would make is I think business model transitions are harder than technology transitions. I think it was harder for most on-premises software companies to move to ratable subscription revenue. Than it …”
Hassabis: General Learning Across Any Domain Is The Defining Feature of AGI
“Learning is a critical part of agenda of AGI. It's actually almost the defining feature. When we say general, we mean general learning. Can it learn new knowledge and can it learn across any domain? That's the general part. So for me, learning is synonymous wi…”
Hassabis: World models are essential for AGI and long-term planning
“And then, of course, that would be, I think, essential for a GI, because that would allow these systems to plan, long-term plan, in the real world over, perhaps, very long time horizons, which, of course, we as humans can do, you know I'll spend four years get…”