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 734 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
Agrawal: Frontier AI Training Data Is Saturated, Shifting the War to Talent
“The ingredients of call it being on the frontier are compute. CapEx, we spoke about it. Data which is sort of access accessible to everybody. We are saturated on, on, on the training data that exists. Talent. This is where the war is now.”
Maslin: Copy-pasting Chinese manufacturing methods to the US will fail
“If you're just copying and pasting China into the United States and trying to build a manufacturing business around that, you are doomed. You will absolutely fail because when you think about the unit economics, labor costs, Environmental regulations. Permits.…”
Taylor: Nuclear and Heavy Construction Are Insulated from AGI Disruption
“I actually think that if you're looking for an investment that is protected from AGI, it's something like this, where it's extremely physical. You're bending steel, you're stacking blocks, you're building plants, you're laying concrete. It's going to be a long…”
Ward: Startups are either execution-focused or multi-product innovators
“I think there's like two types of startups. There's innovation startups and execution startups. So an execution startup is like a rippling or a MongoDB, where it's just like, if you build a better database or a better HRIS system, you have line of sight to bil…”
Ward: Strong sales teams can mask a lack of product-market fit
“One of the biggest mistakes that startups and CEOs when they launch new products make is they have such a great sales motion. They can push a product out to market that doesn't have a strong product market fit, but it feels like it does initially. And then it'…”
Ward's AI rules: No prompt interfaces, must enable impossible experiences
“What I tell the team about how to build AI products is you got to follow two rules. One is differentiated customer experience that we couldn't do before AI existed. Like you do something for the customer that literally we could not have done a year ago. Just, …”
Ward: AI's biggest feature is debuggability, not intelligence
“I think the biggest feature of AI is not that it's smart it's that it's debuggable.”
Government must subsidize mining prices before private capital will invest
“If the government can support kind of a critical problem here in subsidized pricing and help create a market, right? Like there's no price, there's no market that exists right now here because you don't have the ability to get to critical scale... Basically yo…”
Vasavada: Managers are incentivized to overhire to boost their resumes
“Their LinkedIn's are going to say scaled an organization from 20 people to 200 people. And like people think associate their identities with how big of an organization they have or how big of a of an org they were able to scale and over what time. So there's t…”
Power: Automation's highest leverage is a 10x worker productivity boost
“And that's where we think automation has the highest leverage, not to replace people, but to really give the new workforce 10 times productivity, because we just don't have the people in America anymore to build these, you know, critical hardware for our custo…”
Power: Disconnecting production from design degrades design capability
“When you disconnect production from design, you get very bad at design.”
VCs will block life-changing startup acquisitions that miss fund return thresholds
“I think the truth is that even with your board as a founder, I'm not sure your incentives are completely aligned because it's back to, well, first of all, what is the price on the table, right? Is it a nice two or three X for the VCs on the board or not? Becau…”
Financial rails let Brex monetize 20x better than software like Expensify
“We believe that we monetize 20 times better Then other spend management systems, systems like Expensify or Concur, because we monetize through financial services.”
Kim: Generational tech giants often start with controversial or questionable use cases
“Ultimately, we want to back companies that become generational and become very, very large. And a lot of the origin story of companies can be questioned when looked back in from the history or from sort of current looking back, like Facebook started was hard o…”
Kim: AI app cost structures allow high growth to equal high profitability
“The current dynamic of the cost structure, especially for apps companies, Really favor this equation of gross almost equals profitability. The margin structure of some of these companies that we're investors in is incredible.”
Founders waste valuable time obsessing over burn rate and cost-cutting
“If you're a founder and you spend 10 hours thinking about burn and like cutting costs, that's 10 hours you didn't spend about making product and like making product decisions and making your product better and like growing your company.”
Serious founders should eliminate work-life balance and live at work
“You need to work where you live if you're serious about building the company. Like work-life balance should not be a thing that you think about. Your work should be your life and vice versa. And if that's the case, then the best way to realize this is by liter…”
Sadeghi: Embryo selection can only shift IQ by a few points
“Embryo selection will always be a relative optimization, right? IQ can explain a couple IQ points, right? Height, you can, you know, explain a couple inches, but principally if you, the most important kind of the ingredients for your embryos comes from mom and…”
Traditional T2D3 SaaS growth framework is obsolete in AI era
“The triple, triple, double, double, double, sass thing is like, you throw it out the window, you know, it's like, there's a new, it's like, 10 x.”
Startup founders are inaccurately pitching non-recurring revenue as ARR
“Many, many times I have heard something that is not predictable, that is not recurring, be described as error. And that, you know, historically, you know, traditionally error is, is predictable. It's recurring. It's like signed, you know, contracts, usually th…”
Braswell: Speed is currently the only real moat in AI
“I think it's very unclear right now what Out of all these companies who has a moat, but I do think like speed right now is kind of the only moat that you can have.”
Braswell: Hyper-growth startups operate in chaos; calm suggests lacking PMF
“If you are not, like, feeling like you are running around all the time and everything's on fire, then, like, there's probably not a product market fit. Like, there's probably, it's not, there's not that, the only, like, hyper-growth startups, even the ones tha…”
Peters: As AI makes replication easy, brand is the most powerful moat
“In a world where, you know, AI and vibe coding is only proliferating and making it easier to replicate products, brand is possibly one of the most powerful motes and defensible aspects of a business.”
Venture capitalists achieve better investment performance when treating it as a hobby
“I think that it is a better you'll perform better if it's a hobby, so you know, take that as you wish.”
Garrett: European breakout startups face fewer venture-backed clones than US winners
“When you build a winner, you don't have what you have in the U S which is like, once there's a large, massive winner and outcome in the U S there's like 20 startups that get funded to go compete. Right. And Europe, it's not like that. The capital market is not…”
McCormick: Literary nonfiction and deep journalism are best for early-stage underwriting
“That, like, Jim Simons at Rentac versus, like, Robert Carroll, like, who would be a better earlier stage, early stage investor? And I think it's actually, like, Robert Carroll. Like, I think, like, literary nonfiction done really, really well, or journalism do…”
Ritter: Raw corporate data is worthless compared to captured employee workflows
“Remember when it was just data? You can get access to our data. Well, data is nothing. It's the way your employees are using that data to make value out of it. Guess what models do and good models are going to do it even better.”
Guo: Fans Discovering New Creators Double Spending Rather Than Cannibalize
“I think there's like a common misconception where like if A fan, like a fan spends on one creator and it moves to another creator. They spend less on like the previous one, but we've seen that they just ended up spending like double.”
Socher: AI Coding Agents Benefit Average Programmers More Than Top Coders
“At first, like a coding agent, for instance, right now is actually lifting up the at or below average programmers more and more. Like, if you're a solid programmer, but maybe not the best, like, you can get much more productive with an AI agent that helps you …”
Socher: Unicorn seed rounds merge seed risk with late-stage returns
“As an investor, I have shied away from what I call unicorn seed rounds largely because they merge seed stage risk with late stage returns. And if you understand expected value, that's just not a good place to be as an investor.”
Incumbent software systems will capture AI value through existing data gravity
“So I think if that happens and that's what the world looks like, then a LLM looks kind of like a query processor in a database. You give it context, it does its work, you get the answer out. But the point of leverage is the place with, at which the context Liv…”
AI startups should aggressively raise capital every three to four months
“If you are in AI or AI adjacent, you should be raising every three to four months, in my opinion, and you should be doing it aggressively.”
Suster: Seed funds take 15 years to yield most profits
“A seed fund has a 10 to 12 year maturity, and it's really more like 15 years, and that's just the truth about venture capital. It takes longer for your fund to mature, and you might return your fund in year seven, eight, or nine, but you're yielding most of th…”
Suster: Going public without analyst coverage and institutional support is worse than staying private
“Going public, if you don't have coverage, and you don't have an institutional base that wants to invest in you, like, is worse than being private. Because you now have a fixed market cap that doesn't go up. You don't really have liquidity.”
Walk: Underperforming VC funds take a decade or more to exit
“It takes, you know, a decade or more, you know, barring a real firm, you know, a partnership breakup or you know, something illegal, you know, right? Like it takes a decade or more for a underperforming fund To actually exit the market.”
Hunter Walk: Only 5% of VCs are exceptional and 5% are horrible
“And then you get inside the industry and you realize like, actually it's a single hump. There's five percent of VCs who are really, really good at what they do. And even part of that is self-awareness, right? Oh, I'm best in these situations and this stage of …”
Housel: Extreme founder success is normally paired with undesirable personal traits
“So many, maybe there are exceptions to this, but I think it is the norm that the ridiculously successful, crazy genius entrepreneur also has personality traits that you and I would not want to associate with, probably wouldn't want to work with, wouldn't want …”
Housel: Investors must believe absurd stories because winning startups also sounded ridiculous
“So the fact that we can point out Theranos and FTX and WeWork, I think doesn't do justice to say that, like, you shouldn't believe in ridiculous stories, because the ones that do work out were also ridiculous stories.”
Power: US manufacturing must use software to gain 10x labor advantage over China
“The only way for Americans to be globally competitive with China is Literally just a labor equation, right? If we want to keep high wages in America, but compete globally, then we have to have high wage people that are using software and robotics to get a 10 X…”
Power: Vertically integrated deep tech companies either reach Tesla scale or die
“Large vertically integrated companies have two options. One is basically you do the whole thing and you win and you're like TSMC or like Tesla big, or you like die getting there. And there's no kind of halfway mark.”
Power: Series B is the riskiest stage to underwrite deep tech startups
“The way I would underwrite deep tech is basically like series B is the biggest risk point because you're starting to tip real money at the company. C and D is like what's already working. It's a fucking near monopoly. C to A is like, well, it's insane. And you…”
Power: Deep tech startups cannot pivot rapidly like software companies
“Cause the other thing about deep tech is there is no iteration, honestly. Software companies, you can raise a couple million bucks, have an idea, test an idea, ship something new in four weeks. Whoops. Customer base sucks. Pivot, pivot, pivot, pivot, pivot. Th…”
Power: Post-recession startups succeed due to founder pain thresholds, not asset pricing
“One of the reasons why the best companies have founded directly after recession has nothing to do with like pricing. It's just that you get people like Airbnb or Travis from Uber where they're just like, I have no other option. So my pain threshold is stupidly…”
Kolicich: Big Tech's Low AI Pricing Will Hurt Competing Startups
“Big tech is doing a good job of anchoring the price of all these features pretty low, like co-pilots, 20 dollars a month per developer. It's incredibly cheap relative to the value that it does add. And that really anchors, you know, value in that market pretty…”
Kolicich: Big Tech Subsidizing Open Source AI Is Hard for Startups
“They're really subsidizing the space, a lot of big tech, which I think is positive for open source and positive for users, hard for startups.”
Mendler: Modular ground systems can match large parabolic dishes' performance
“If we want to accomplish serving the Stanford dish while serving a lot of other things, we probably would build something that's more modular. They could add up in the sum of its parts to the kind of performance that's required for that operation. But in the m…”
Beck: Space startups linger as zombie companies longer than other industries
“The funny thing is, like, space companies are really hard to kill. Like, the amount of space companies that should be dead, ah, and not, the sort of zombie companies is quite, quite incredible. It's not like other industries where, you know, you go to die, you…”
Beck: Space startups fail because long-term hardware reliability is far harder than reaching orbit
“The most amount of failure I'd say is because people don't build reliable stuff. And it's easy to kind of rush it and throw something on orbit. But I think the distance between putting something on orbit and putting something on an orbit that can last 12 years…”