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 1,960 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
Garg: Low farmgate prices reflect value addition, not farmer exploitation
“There's many times you would be hearing this argument in our country that the farmers are looted they get for the papaya or for the apple they produce, they get only 10% or 20% or 30% of the final price which is paid, right? This is a very short-sighted argume…”
Ferrari: Finding product-market fit is overwhelmingly driven by luck for most founders
“Our hypothesis was that to find product market fit with something very new, Most of it was luck. We felt talent and hard work came nowhere near guaranteeing you would. I mean, maybe, maybe there are some incredibly rare talents. Some people can see it, but I'm…”
The intellectual NGO ecosystem relies on a business model glorifying poverty
“If you, if your you know, great economists are basically glorifying poverty, then don't complain that you get poverty, but because in each one of these things, an economy arises out of it, you know, so you will get an, you know an entire povertarian system wil…”
Western billionaire foundations fund sensationalized health studies to influence Indian policy
“One of these billionaire foundations will go and fund these studies. Then those studies will show either nutrition, vitamin D deficiency, whatever, whatever it is, the latest thing that you want to sell to India. And then it will be, then next stage will be, i…”
Shenoy: Financial Regulators Should Only Penalize Influencers for Provably Malicious Actions
“I think the action needs to be justifiable justifiably malicious or provably malicious.”
Menon: Delivering every grocery order in 10 minutes is structurally unprofitable
“Nobody can profitably deliver every order in 10 minutes. It's not possible. The network of dark stores you'll have to set up to do that is not profitable. So you will never be able to build a good unit economics based on only 10 minutes, which means all orders…”
Mehta: Brands can move from premium to mass, but never mass to premium
“That's the third immutable law of brand building that you can only build a premium brand and make it go mass. You can never build a mass brand and make it go premium. There is not even one example in the world where someone built a brand, which was first adopt…”
In late-stage investing, fundamentals follow capital, not the other way around
“Capital doesn't follow fundamentals. Fundamentals follow capital.”
Lunia: India has a lower-middle-class boom, not a middle-class boom
“India doesn't have a middle class boom. It, however, has a huge lower middle class boom. Right. A huge number of people entering at 20, 25,000 rupees monthly income.”
Reddy: India backs me-too copycats instead of novel behavior creation
“And it's because of new behavior creation that incredibly long, large long-term value gets created in the world. And India is almost the antithesis of it. All the large-term value you're trying to create is in Me Too's. Right? By pumping more money and trying …”
Sankar: Selling to personal networks creates fake traction because people want to help
“I believe, most people love founding stories. Most people want to help. And most people will find a way if you speak with them and tell them this is the problem you're trying to solve, they'll find a way to help you. And that doesn't mean that they are going t…”
Sankar: AI-native companies must foundationally redesign hierarchies and processes
“You have to like leveling structures, like you have to foundationally reimagine how you operate as a company to be truly AI native or on the frontier versus I think The other option of it is you could say, hey, this is how I do things today. I'm gonna go slap …”
Sankar: Settling for average PMF stalls startups at $5M to $7M
“You either never find product market fit or you find product market fit, but it's an average product market fit. And so you peter out at five million or seven million because you can't scale beyond that.”
Sankar: Frontier tech salespeople must be builders to hold customer credibility
“Well, I actually think they do. I think if you're trying to sell to the frontier what, how are you going to have a conversation with a customer who's talking to you about, you know, the difference between model gateways and three harnesses and memory systems, …”
Gupta: Three early customers are a startup 'kiss of death'
“If you know you'll have three customers, I call it as a kiss of death. Because anybody like who's 42, 45, who is in an industry experience of 20 years, you'll be able to get three customers, you know, I know.”
Tandon: Healthcare's core bottleneck is administrative overhead, not clinical tools
“The clinicians had the tools they needed for the average patient to get the job done. The real problem was administrative. It was in how they ran their telehealth operations and how did, and how they did their billing and how they, you know, fought insurance c…”
Kompalli: Indian universities outperform international brands in India's online education market
“Strong Indian universities do so much better, ah, because of like local, ah, demand and acceptance.”
Jain: Indian SaaS startups cannot enter the US by just hiring reps
“And a lot of times what founders end up doing is they say, okay, it's so easy. I just hired two sales rep in the US. And I'm like, never, it's never, ever going to work because it's not just about your sales. Your entire company is meant and built For Indian n…”
Goel: Multimodal AI Models Degrade in Intelligence Versus Text-Only Models
“They tend to degrade in terms of their intelligence when you try to build models that can do multimodal reasoning. Like text only models tend to be pretty smart, but when you do multimodal, they tend to get a little bit less smart.”
Krishnan: End-to-End Outcome Startups Benefit from Smarter Frontier Models
“Now, when you're doing this kind of end-to-end thing, you are less likely to get disrupted. In fact, if you know, if anything, if the if the cloud models or codex or even Gemini models or any of these other models get a lot better at coding, these people only …”
Krishnan: Rapid growth and 10x product improvements trump AI defensibility
“The number one thing before you know, you get into defensibilities, make sure that whatever it is you're building is is was impossible to build four years ago. Make sure, you know, it is a 10 X or a hundred X better than the status quo. Make sure that it has a…”
Krishnan: AI Accelerates Execution but Not Fundamental Innovation
“Yes, while this has done a lot to the execution part, it does not really seem to have speeded up the innovation and the invention part, because theoretically, imagine you speed up everything, then who cares? There are 10 times the number of venture-funded star…”
Krishnan: Purely model-generated synthetic data hits a wall without human feedback
“There is also, I think, limits to how much value can be added there because ultimately there is just not too much new information. If you're telling the model itself to kind of generate, yeah, There is a, I mean, there's all sorts of things about how beyond th…”
Krishnan: Meaningful services provide AI startups a moat against model advances
“Having having the need for a meaningful services component does, for example I mean, it serves a useful purpose also today, right? Like in a manner which might not have been valued as much a few years ago, which is that It absolutely gives you a certain moat a…”
Krishnan: Startups acting as band-aids for model flaws won't survive upgrades
“I, even a couple of years ago, I always thought certain categories of startups, which were doing some almost like a bandaid on what is today's a hole in today's model is really not going to survive because the next version doesn't need that bandaid basically. …”
Most AI startup moats will not be technical, excluding foundation models
“Most of the mode is not going to be technical unless you're a foundation model company or a world model. It's not technical.”
Foundation models struggle with multi-stakeholder enterprise workflows due to missing memory
“The foundation models aren't built for that because they don't really understand persona. They don't have memory. You know, you think they don't have memory of multiple complex and the interdependencies between those. So I actually am seeing founders, some of …”
Penmetsa: AI products see faster customer adoption but face much higher churn
“The pace at which you can get a buyer and potentially even churn is much higher, right? Because the AI products, one of the benefits of that is it is easier to build, but also I think it is easier to put it in and replace it, right?”
Penmetsa: Owning the stack between AI models and UI creates defensibility
“Companies really need to think about how are you building that full stack between the model layer and the UI. Like, you need to own that entire thing, because if you own the entire thing, at least in some verticals, It is a lot of work to rip you out. And seco…”
Venture returns depend on company quality, exit ownership, and entry price
“In our business, I think three things matter in that order. I think number one, it's a parallel business. So you've got to be in the best companies. A few companies end up driving the returns, whether it's for your fund or for the industry, and you've got to b…”
Dalal: First hardware products shouldn't try to create new markets
“And one of our core belief and thesis is that when building consumer hardware, you do not want to, with your first product, create a new market.”
Sanyal: LLM evaluation does not require general-purpose reasoning models
“We feel like if you really focus on the fundamental problem that, hey, evaluations for language models is a very task-specific, constrained problem, and it does not require you to use you know, general purpose reasoning for that, and then there's fine-tuning y…”
Sanyal: Software and code are commoditized into tokens by fast AI generation
“In an era where you can build an, you know, a website and a web application with a Postgres database behind in 30 seconds, Software is just commoditized. Software is just tokens. Code is tokens.”
Senpaty: Thoroughly underwritten education loans behave like secured debt
“So somebody who does their job well in terms of underwriting these institutes or the courses, the end uses and the students will have a very low NPA. It will behave like a secured product is what our understanding.”
Tewari: Building something massive requires similar effort to building something moderately big
“The effort required for doing something decently big to doing something magnanimously big is not that different.”
Soni: MarTech Customer Retention Is Structurally Subpar
“So the retention in MarTech is always subpar. That is a well-known fact, which creates challenges in fundraisers and everything else.”
Soni: Adobe and Microsoft avoid acquisitions below $1B enterprise value
“The likes of Adobe, Microsoft, those companies, right? Those two They don't move as much until it's like a billion in enterprise value, right? Because it's just too small for them. So players like us are not attractive to them in general.”
Subramanian: AI driving software costs to zero eliminates pure technology moats
“Sustainability has to come through some sort of business mode and not through just technology mode alone. Because there's nothing called a technology mode and that's gone for everybody, right? It's become a zero sum game for all of us, and when software become…”
Siddharth: Coding, Tool Use, Reasoning, and Multimodality Unlock Superintelligence
“If you can solve coding, tool use, reasoning, and multimodality, you have the keys to superintelligence. Those are the building blocks. If you can solve those four, you can automate anything that a human does in front of a computer, and it'll give humans super…”
Siddharth: Humans Should Never Create V1 of Knowledge Work
“Basically the agents create V-one of the work. Humans are verifying and iterating. Humans never create V-one.”
Siddharth: Agentic AI models will eliminate the need for intermediate SaaS
“So on one hand, the models are becoming agentic. And in theory, if the models have access to the right sources of data, You kind of don't need software in the middle. The models can do it end to end.”
Siddharth: Basic Workflows Suit Fine-Tuning; High-Value Use Cases Need Frontier Models
“Step one should be to, like, just do it with a frontier model and see how well it does. And most times you'll see that you don't have to touch the weights. But there are certain workflows like customer support like an invoice to pay system, some of these more …”
Saxena: The classic T2D3 SaaS growth playbook is obsolete in the AI era
“There used to be like, Battery and Neeraj had mentioned whatever the three triples, two doubles kind of a thing triple, triple, double, double, double, or whatever that, whatever the norm used to be that playbook's out of the window, right? Like right now with…”
Saxena: AI startups are one Anthropic press release away from obsolescence
“Now, in the AI wave, I feel if you're heads down and you're focused, you don't know what's going to come hit you from the left or right. You are one anthropic press release away from not having a business ahead of you.”
Saxena: Traditional Enterprise Rep Models Cannot Reach $100M ARR in Two Years
“Because the traditional model of, like, enterprise reps doesn't get you to a hundred million ARR in two years. You have to do something drastically different for that, right? In fact, a lot of these hot AI startups aren't doing the traditional model. Be it for…”
Saxena: In AI, Only 20% of Startup Playbooks Are Replicable
“So yes, there's, so instead of like, 80% of the playbook being replicable, now in the AI world, only 20% of the playbook is replicable. So I don't know if being a repeat founder has that much of an advantage in this narrow time frame where it's AI.”
Reddy: Enterprise Self-Serve SaaS Is A Myth; Executives Delegate To Agencies
“What happens in large enterprises is self-serve is a little bit of a myth. Because no senior guy in a large enterprise No one wants to sit in a two hour training and learn a UI. So you, unfortunately, you get delegated out to an agency or to a, you know, a sys…”