Mamoon Hamid: AI - Where Value Accrues, Startups vs Incumbents & Scaling Laws | E1217
Mamoon Hamid (guest), Harry Stebbings (host)
In this episode of The Twenty Minute VC, featuring Mamoon Hamid and Harry Stebbings, Mamoon Hamid: AI - Where Value Accrues, Startups vs Incumbents & Scaling Laws | E1217 explores mamoon Hamid on AI’s boom, venture discipline, and market creation Mamoon Hamid argues we’re in a once-in-a-lifetime AI supercycle where value will accrue largely at the application layer, especially in software that amplifies scarce, high-value workers like doctors, lawyers, and developers.
Mamoon Hamid on AI’s boom, venture discipline, and market creation
Mamoon Hamid argues we’re in a once-in-a-lifetime AI supercycle where value will accrue largely at the application layer, especially in software that amplifies scarce, high-value workers like doctors, lawyers, and developers.
He maintains that early-stage AI investing is fundamentally the same craft as classic SaaS: back exceptional, deeply technical, product-obsessed founders in big or category-creating markets, while staying disciplined on ownership and pricing despite hype.
Hamid is skeptical of overinvestment in AI middleware and pure LLM providers in the near term, but believes massive CapEx will be justified as AI shifts spend from software licenses to labor substitution and productivity, unlocking trillions in tech value.
On venture practice, he stresses concentration, thoughtful reserves, non-formulaic deployment, reputation-driven access, and founder selection over data platforms or voting structures, while candidly discussing mistakes in reserves, lending bets, and over-believing for too long.
Key Takeaways
AI value will concentrate in vertical, application-layer products that augment scarce talent.
Hamid focuses on tools for doctors, lawyers, and developers, arguing AI that directly multiplies expensive human labor (e. ...
Get the full analysis with uListen AI
Founders and product quality still matter more than model access in AI startups.
He sees differentiation coming from deeply technical founders paired with domain experts, and from tuning models for near-perfect output in critical domains (e. ...
Get the full analysis with uListen AI
Venture investors must stay price-disciplined, using “YOLO” bets sparingly.
Hamid acknowledges massive pre-product rounds at extreme valuations but insists these should be rare exceptions; the core business still requires buying 15–20% ownership at sane prices and reserving 40% of capital for follow-ons, or the fund math breaks.
Get the full analysis with uListen AI
Much AI middleware and tooling may be overfunded and fleeting in value.
He believes the “middle layer” (vector DBs, fine-tuning infrastructure, orchestration) is seeing heavy investment, but rapid technical change and potential commoditization mean much of the value there may not endure compared to focused applications.
Get the full analysis with uListen AI
LLM platforms will likely become good businesses, but not quickly or easily.
While today’s LLM providers have weak margins and intense price competition, Hamid draws an analogy to AWS and cloud: as GPU performance improves and scale grows, selling “electricity and compute” can become highly profitable over time, even if token prices keep collapsing.
Get the full analysis with uListen AI
Great category-creating products can justify high early valuations if engagement is undeniable.
In deals like Slack and Figma, Hamid paid seemingly extreme multiples (e. ...
Get the full analysis with uListen AI
Venture is a conviction craft, not a data or committee-driven optimization problem.
He dismisses proprietary data platforms and voting-based investment processes; at Kleiner, any partner can lead a deal without formal votes, decisions are driven by individual conviction stress-tested in small-room debate, and founder demand for you as a partner is the true edge.
Get the full analysis with uListen AI
Notable Quotes
“We’re in the midst of a supercycle like none we’ve seen before… this time feels like the rise of the internet multiplied by 10.”
— Mamoon Hamid
“We took the top 20 jobs in the US and who makes the most? It’s doctors, lawyers, and developers. How do we help supercharge these people?”
— Mamoon Hamid
“I love products that create markets. They get to create the playing field, they play on the playing field, and they win the game.”
— Mamoon Hamid
“A lot of firms think that information is knowledge and knowledge is arbitrage. But when everyone has the same knowledge, it’s no longer arbitrage.”
— Mamoon Hamid
“Venture is a grind. It’s glamorous… except it’s not glamorous.”
— Mamoon Hamid
Questions Answered in This Episode
If AI is primarily about replacing or amplifying labor, how should startups think about pricing and business models beyond classic seat-based SaaS?
Mamoon Hamid argues we’re in a once-in-a-lifetime AI supercycle where value will accrue largely at the application layer, especially in software that amplifies scarce, high-value workers like doctors, lawyers, and developers.
Get the full analysis with uListen AI
Given the overinvestment risk in AI middleware, what characteristics would convince you that a middleware/tooling company has durable, defensible value?
He maintains that early-stage AI investing is fundamentally the same craft as classic SaaS: back exceptional, deeply technical, product-obsessed founders in big or category-creating markets, while staying disciplined on ownership and pricing despite hype.
Get the full analysis with uListen AI
How can founders realistically assess when a high valuation today may damage their ability to raise or pivot later, and how should they push back in the moment?
Hamid is skeptical of overinvestment in AI middleware and pure LLM providers in the near term, but believes massive CapEx will be justified as AI shifts spend from software licenses to labor substitution and productivity, unlocking trillions in tech value.
Get the full analysis with uListen AI
In a world where incumbents control most frontier models, what concrete advantages can early-stage startups still exploit to build enduring, independent companies?
On venture practice, he stresses concentration, thoughtful reserves, non-formulaic deployment, reputation-driven access, and founder selection over data platforms or voting structures, while candidly discussing mistakes in reserves, lending bets, and over-believing for too long.
Get the full analysis with uListen AI
How should VCs and founders adapt reserves, follow-on strategy, and capital efficiency if IPO and M&A markets remain muted longer than expected?
Get the full analysis with uListen AI
Transcript Preview
(instrumental music) I love products that create markets. Slack created a market. Figma created a market. They get to create the playing field, they play on the playing field, and they win the game. We've invested in a lot of application-layer companies. We took, actually, the top 20 jobs in the US and who makes the most? And it's doctors, it's lawyers, and it's developers. How do we help supercharge these people who are highly scarce, highly skilled and we're not producing enough of them?
Ready to go? (instrumental music) Mahmood, I am so excited for this, man. I can't believe it. You just reminded me that SaaStr nine years ago was our first show. Thank you so much for joining me today.
Thank you so much, Harry, for having me. It's so great to be here.
Listen, I wanted to start... I started an LP update the other day that I did with, "It is the most exciting time to be in venture. It's also the hardest time to be in venture." Would you agree with that statement, high level?
It is the most exciting time to be alive. It is. We're in the midst of a, a supercycle like none we've seen before. Uh, the AI supercycle, as you know. And it reminds me of the time when I first came to Silicon Valley in 1997. I was 19 years old, and it was all just roses all around me. It was the rise of the internet. And this time feels much like it, multiplied by 10. And that, obviously, puts us in an interesting spot as venture investors who get to invest into this cycle, this, this sort of massive tidal wave of change that's coming. And yeah, the world's not going to be the same anymore.
It's not. Uh, the thing that's seismically different for me when I look at the two, and, uh, I don't mean to age you, I was four in that kind of period.
Yeah.
We didn't have the incumbents spending $100 billion on frontier models. Larry Ellison said the other day, "It's going to be $100 billion to enter the frontier model race." And you're looking at that going, "Christ, that is a different level of incumbent spend than we've ever seen before." How do we think about that, as it is a fundamentally different addition?
Yeah, we have some very strong incumbents, Google, Microsoft, Amazon, Meta, Oracle-
(laughs) .
... who can all spend hundreds of billions on these front-end models. So you're absolutely right, and Larry's absolutely right, of course.
Does that make it harder for us as venture investors? With the rise of corporate investors who maybe have d- different motives or different incentive structures, does that make it harder for us?
It doesn't, uh, because I think the opportunity is still in front of us. I think the, uh... we can talk a lot more about AI, but just, it, there are so many things to build on top of this infrastructure, all these frontier models, that is going to create so many trillions of value over the next decade.
Install uListen to search the full transcript and get AI-powered insights
Get Full TranscriptGet more from every podcast
AI summaries, searchable transcripts, and fact-checking. Free forever.
Add to Chrome