Apr 21, 2023 · 59m · news
Tomasz Tunguz: How I Raised $230M; ChatGPT vs. Google; How LLMs Work; Trump vs DeSantis | E1004 · 20VC with Harry Stebbings
0:00 / 0:00
▶ Watch on YouTube →
gold bands on the timeline = statements, start to end. Hover to read, click to jump. CC turns on captions
In this episode of 20VC, host Harry Stebbings interviews Tomasz Tunguz on his transition from Redpoint to launch Theory Ventures, detailing his tactical playbook for raising a $230 million fund and his deep strategic insights on enterprise AI, startup-incumbent dynamics, and macroeconomic trends.
How this conversation actually went
Every chapter scored 0–10 on four independent dynamics. Hover any point for the reasoning behind the score. Harry holds 27.6% of the talking time here. How this is scored →
Harry as informed peer 5.6 Guest teaching 4.8 Guest disagreement 1.2 Harry pushing back 2.6
speaking balance: gold is Harry, purple is the guest (3 minute bins)
Sharpest disagreement ▶ 39:40 Reframe on wealth inequality and historical power laws
When Harry raises concerns about AI worsening wealth inequality, Tomasz dismisses the novelty of the premise, stating that it touches on politics and is a historical constant seen in telecommunications and railroads.
Hardest push from Harry ▶ 19:25 Harry challenges exit market projectionsHarry explicitly questions the validity of performing exit market analysis given how drastically multiples fluctuate over 7 to 10 year venture horizons.
Biggest teaching moment ▶ 36:30 The Milan bicycle experiment on explanatory depthTomasz educates Harry on buyer psychology by citing a study where only 2 out of 100 people could correctly draw a functional bicycle, illustrating that enterprise buyers do not need to understand AI mechanics to adopt it.
Harry holds his own ▶ 17:07 Harry counters with his own Monte Carlo numbersHarry demonstrates his own technical rigor by matching Tomasz's quantitative frame, citing specific Monte Carlo findings that 23 portfolio companies yield over 80 percent of diversification benefits.
the scores for every segment, with the reasoning behind each
| Chapter | Topic | Harry as informed peer | Guest teaching | Guest disagreement | Harry pushing back | Why |
|---|---|---|---|---|---|---|
| Raising a $230 Million Fund in a Hard Market | 6 | 3 | 1 | 2 | Harry demonstrates strong venture fundraising knowledge by offering operational advice, asking detailed questions about LP pipeline mechanics, and sharing his own LP intro strategy. Tomasz answers collaboratively and describes his software-sales approach to LP fundraising. | |
| Designing the Optimal Portfolio and Investment Strategy | 7 | 4 | 1 | 3 | Harry demonstrates deep quantitative understanding of portfolio construction, citing his own Monte Carlo simulations on diversification benefits. Tomasz responds with detailed frameworks on Fermization and lessons from leading historical insider rounds like Snowflake. | |
| The Future of AI: General Models vs. Application Layer Moats | 5 | 5 | 1 | 2 | Harry asks sharp structural questions about cloud architectures and application layer value capture. Tomasz provides an insightful comparison showing the market capitalization parity between three cloud providers and top application layer SaaS companies. | |
| AI-Driven Economic Prosperity and Responsive Regulation | 5 | 6 | 2 | 3 | Harry raises concerns regarding enterprise buyer sophistication in Europe and potential macro wealth inequality. Tomasz educates on technology comprehension using psychological studies and illustrates regulatory iteration using aviation history. | |
| Startups vs. Incumbents and Google's Innovator's Dilemma | 5 | 6 | 1 | 3 | Harry presses Tomasz on Google's strategic mistakes despite acquiring DeepMind. Tomasz delivers a nuanced explanation of the innovator's dilemma and explains emergent LLM behavior using a memorizing versus swimming analogy. |
Statements from this episode (29)
Tunguz: Concentration near power law apex yields higher VC returns
“A venture capital, the industry is governed by a power law. And the more dollars you can have closer to the y-axis, so to speak, on the power law, the better your returns will be.”
Tunguz ran 150 LP meetings to raise $230M Theory Ventures fund
“It took about a 150 LP meetings. So the fundraising market was a very challenging one over the last couple of months, I'd say, but it took about a 150 meeting about a 150 LPs. And the way I thought about it was, I mean, you know me, I thought about it just lik…”
Half of Theory Ventures' $230M fund came from new LP relationships
“So about 50% of the capital was for new relationships, about half of the capital.”
Tunguz capped Theory Ventures' largest LP check at 12% of fund
“I think the largest LP is no more than 12% of the fund.”
Tunguz: Single fund closes are vanity metrics for VCs
“The purpose of a close date is just to drive people in unison to a cadence that you're trying to set. There's nothing magical about it. There's nothing terrible about having multiple closes. It's just a way of organizing a particular process. And that close da…”
Tunguz: Successful fundraising requires convincing investors of round's inevitability
“When you're fundraising, what you want to convince people of is inevitability. That the company or the fundraising round, its conclusion, its positive conclusion is inevitable.”
Up to a third of Theory Ventures LPs committed before meeting in person
“Maybe I want to say like a quarter to a third of the LPs I only met after they committed in person. And that's probably an overhang from COVID where a lot of funds were raised entirely virtually and people are comfortable.”
Tunguz: Theory Ventures will back 12-15 companies, concentrating 40-50% in top 3
“It's about 12 to 15 portfolio companies. You have significant concentrations, so you probably have like 40 to 50% of the fund in the top three holdings, maybe more.”
Tunguz: Theory Ventures will write $8M-$12M initial checks
“Yeah, it's about eight to 12 initially.”
Tunguz: Software forward revenue multiples dropped from 40x during QE to ~6x
“Historical forward multiple is about five X, let's just say, right? It's a little higher than that's about 5.5. And in the heyday of quantitative easing, the top quartile companies are trading at 40 times. And so you can't go, and today it's about maybe six.”
Tunguz: Startup funding conversion is 60% Seed-to-Series A and 50% Series A-to-B
“The base rate for hiring, for raising a series A from seed is about 60%. Then they need to raise a series B. Base rate is 50%.”
Tunguz: Investors must be explicit buyers or sellers; middle positions reflect ignorance
“With every stock position that you have, if you're any kind of investor, you either, you should either be a buyer or a seller of that position. Right. And if you're in the middle, you probably don't know enough about a business”
Tunguz: Snowflake struggled to raise external capital at Series C
“The company was burning a ton of capital and couldn't raise money from the outside. And it was the insiders that stepped up and led that round because they believed in the business.”
Tunguz: Web2 cloud infrastructure and application layers both have $2.1T market cap
“So in web two, if you take the top three clouds and you look at their market cap, so AWS, GCP, and Azure, it's about a 2.1 trillion dollar market cap just for the cloud businesses. And then if you take the top 100 publicly traded cloud companies, both on B to …”
Tunguz: VC success odds are higher at AI application layer than infrastructure
“And so if the analogy holds, then you really, you know, as an investor, it's the odds of success are going to be significantly higher at the application layer because the diversity of needs there is greater.”
Tunguz: Healthcare and finance enterprise AI will stay on-premises
“I think if you're in. Like finance or healthcare, you'd probably be completely on-prem for the foreseeable future.”
Tunguz: Enterprise buyers in early markets prefer bundled software over best-of-breed
“So my learning has been that in early markets, people want bundling. They want bundling because they don't yet understand the tech, the technology moving so fast that most people don't really understand it end to end. But they want the technology to solve a pr…”
Tunguz: AI Will Generate 70% to 80% of Code Within 10 Years
“I think it'll probably be 70 to 80%, and the reason I say that, I bet that 40%, a lot of it is what's called boilerplate code.”
Tunguz: Government Regulation Systematically Benefits Market Incumbents
“I think regulation on the whole, as a, one, it benefits incumbents because the cost of adhering to regulations are significant.”
Tunguz: Going Public with $100M Revenue Costs $15M in Year One
“Today, if you have a hundred million in revenue, it costs, then it costs you fifteen million dollars in your first year to go public.”
Tunguz: Superior execution remains the primary competitive moat for software startups
“And I think the answer is the one that it's always been, which is better execution is the moat. If you can build a better CRM and get it into market, you can win, right?”
Tunguz: Chat interfaces will replace traditional web search for many use cases
“I didn't believe that chat would replace search, but I think for many use cases, it will.”
Tunguz: AI model sophistication is far from reaching an S-curve plateau
“We're on the steep part of this geometric curve for the sophistication of these models, and at some point, it will make the shelf of an S, but we're not, it doesn't feel like we're anywhere close.”
Tunguz: Macroeconomic conditions will deteriorate by end of 2023
“I think we will probably be in a worse place by the end of 23.”
Tunguz: Adobe's generative AI capabilities are underrated relative to competitors
“Adobe I think is doesn't have doesn't have the recognition recognition it deserves. When it comes to using generative, I think about like the applications in Photoshop, they launched a product called Firefly. I think they're right there.”
Tunguz: Startups with fanatical users in early markets surprise on upside
“The thing that I've learned is that the startups are the ones who create the markets. And so if you have a rabid user base in a really early market, it will most of the time surprise you on the upside.”
Tunguz: DeSantis will win the Republican primary over Trump
“I don't think so. I think, I bet DeSantis wins. I think it will be tough for him to circumnavigate all the legal troubles, and I wonder if the RNC doesn't get involved.”
Tunguz: US entitlement spending projected to hit 95% of tax receipts
“Entitlement spending in the U.S. Over the next 10 years is projected to consume something like 95% of tax receipts, and so that can't be.”
Tunguz: Looker was his top cash-generative DPI investment
“It was Looker. And you know, the story there was in 2012, Redshift was the fastest growing product inside of AWS, and Tableau was the dominant BI product. And there was a thesis that there would be a new BI product that would be architecture for the cloud.”