Mar 19, 2026 · 1h 17m · 20vc
NVIDIA Predicts $1TRN in Revenue: Everything You Need to Know From GTC & Anduril Lands $20B Contract · 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 the 20VC podcast, host Harry Stebbings, along with veteran investors Jason Lemkin and Rory O'Driscoll, analyzes major shifts in technology and capital markets, including Nvidia's valuation limits, AI's disruption of corporate workforce structures, venture capital fund discipline, and the realities of executive leadership.
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 8.4% of the talking time here. How this is scored →
Harry as informed peer 3.3 Guest teaching 4.9 Guest disagreement 3.0 Harry pushing back 2.4
speaking balance: gold is Harry, purple is the guest (3 minute bins)
Sharpest disagreement ▶ 1:04:54 Rory bluntly tells Harry to halve his fund size
When Harry argues that massive growth funds must deploy capital into Travis Kalanick, Rory directly attacks the premise, telling Harry to halve his fund size if that is his investment logic.
Hardest push from Harry ▶ 1:05:24 Harry hits back on fund size and AUM strategyHarry immediately fires back at Rory's fund discipline lecture by pointing out that Rory's strategy simply means not playing the AUM game.
Biggest teaching moment ▶ 15:10 Rory corrects Jason on Meta's real cash flows vs compute spendRory breaks down Meta's true financial picture, correcting Jason's assertion about operating margins by pointing out that after accounting for massive Nvidia capex, Meta's free cash flow is near zero.
Harry holds his own ▶ 1:04:01 Harry demonstrates growth fund capital deployment mechanicsHarry articulates the structural imperatives of multi-billion dollar growth funds like General Catalyst or Coatue, explaining why deploying $200M+ checks into elite founders is a mathematical necessity.
the scores for every segment, with the reasoning behind each
| Chapter | Topic | Harry as informed peer | Guest teaching | Guest disagreement | Harry pushing back | Why |
|---|---|---|---|---|---|---|
| GTC Key Takeaways and Nvidia's Massive CapEx Forecast | 2 | 4 | 2 | 1 | Harry sets up the GTC topic asking about data centers in space. Rory breaks down the underlying math of Nvidia's $1T forecast, showing how analyst expectations already accounted for the growth. | |
| Will NVIDIA Reach a $10 Trillion Valuation? | 3 | 5 | 3 | 2 | Harry asks if Nvidia will hit a $10T market cap, but Jason corrects the framing to cumulative revenue rather than market cap. Rory highlights the risk of falling token prices offsetting volume growth. | |
| Analyzing Atlassian and Meta's Workforce Reductions | 3 | 4 | 2 | 1 | Harry introduces the Atlassian and Meta layoff numbers with exact figures. Jason explains that these mega-cap layoffs are strategic restructurings for the AI era rather than cash-flow desperation. | |
| The Five Categories of Modern Layoffs | 0 | 6 | 3 | 0 | The host is silent in this monologue segment. Rory reframes the layoff discussion into five clear categories and corrects Jason on Meta's free cash flow accounting relative to massive Nvidia capex spend. | |
| The Rise of the Agentic Deployment Expert (ADE) | 3 | 5 | 2 | 1 | Harry asks targeted questions about modern hiring profiles. Jason delivers an impassioned rant coining the concept of the Agentic Deployment Expert (ADE). | |
| How Generalists and Non-Technical Leaders Win with AI | 2 | 4 | 1 | 1 | Harry interjects with a brief comment about their main investor sitting on the bleeding edge. Jason outlines the progression from prompt engineering to generalist software deployment in AI. | |
| Inside Anduril's $20 Billion Prime Defense Contract | 4 | 5 | 1 | 1 | Harry provides exact contractual details on Anduril's $20B defense contract. Rory explains how Anduril achieved prime contractor status by consolidating fragmented software layers. | |
| The Valuation Trap and Debating the Death of Small TAM Bets | 5 | 4 | 5 | 5 | Jason claims small TAM seed bets are dead. Harry pushes back by asking if the best markets start small and naming specific startup categories, while Rory warns against power law myopia. | |
| How AI Expands TAM and the Cycle of Excess Capital | 5 | 5 | 4 | 5 | Harry challenges Jason's anti-small TAM stance using Replit's early seed stage as a counterexample. Rory cites Barton Biggs to illustrate how excess capital destroys returns. | |
| Travis Kalanick's Atoms: Why Wheels Beat Legs in Robotics | 3 | 4 | 2 | 1 | Harry brings up Travis Kalanick's return with Atoms after eight years in stealth. Rory validates Kalanick's robotics thesis that wheels beat humanoid legs for industrial use cases. | |
| Uber's Trillion-Dollar What-If and the Steve Jobs Playbook | 3 | 5 | 6 | 2 | Jason asserts Uber would be a trillion-dollar company today had Kalanick remained CEO. Rory forcefully disagrees, detailing the public market pressure for profitability and cash flow constraints. | |
| The Brutal Reality of Swapping Out Founders | 4 | 6 | 3 | 6 | Harry repeatedly presses Rory to answer whether he would have voted to remove Kalanick. Rory details the severe risks and strict criteria required for board-led founder replacements. | |
| Fund Size Strategy and Betting on Travis Kalanick | 7 | 6 | 6 | 7 | Harry takes a firm stance that multi-billion growth funds must back Kalanick. Rory sharply criticizes this check-writing logic, telling Harry to halve his fund size, to which Harry counters with a sharp remark on AUM strategy. | |
| Adobe CEO Resignation and the Reality of Public Company Leadership | 3 | 5 | 3 | 2 | Harry introduces the Adobe earnings beat and CEO departure. Rory corrects the timeline, noting the combined announcement triggered the stock decline. | |
| Evaluating Adobe's Growth and AI Disruption Risks | 2 | 6 | 2 | 1 | Harry asks for a five-year outlook on Adobe. Rory contrasts Intuit's durable tax workflow moat against Adobe's vulnerability to generative AI workflow disruption. |
Statements from this episode (35)
Lemkin: Uber would be worth $1 trillion today under Travis Kalanick
“I think Travis Uber would be a trillion dollar company today because it would be five years ahead of where it is today.”
Lemkin: Winning with AI agents in Q2 2026 requires zero technical skill
“You do not need to be technical to win with AI agents in Q two of 26. You do not need to be even one percent technical.”
O'Driscoll: Excess capital will destroy even the best investment opportunities
“There's no investment opportunity so good that excess capital won't destroy it.”
Lemkin: OpenAI is struggling and facing an internal code red
“You can smell open AI struggling right now. You can just smell it. You can smell the code red and the fact that they said yesterday, we have to concentrate on enterprise. We have to stop side projects.”
O'Driscoll: Nvidia's $1T soundbite merely matches analyst forecasts
“So it turns out if you add 504 hundred and then you apply a salesman's roundup, you get to a trillion dollars. Another way of translating that wonderful, amazing number of a trillion dollars is the analyst forecasts for the next two or three years look roughly…”
Lemkin: Global AI inference volume will grow 1,000x in five years
“Yeah, I mean, I think it's got to be three orders of magnitude more inference we run in the next five years.”
O'Driscoll: 30% chance tech CapEx growth slows within five years
“There is at least some probability, pick a number, 30%, that it doesn't happen this way.”
Lemkin: Nvidia will reach $10 trillion cumulative bookings in 5-7 years
“I think there's a high degree of confidence that most of the roads lead back to Nvidia reaching 10 trillion of cumulative, cumulative bookings over the next five to seven years.”
Lemkin: Meta and Atlassian layoffs are voluntary, not financially forced
“Neither of these companies has to lay off anybody. Okay. Atlassian has substantial free cashflow meta. I don't know. They're in the, their free cashflow is dip, but it's still like in the mid forties, right? They don't have to do this.”
Lemkin: Existing SaaS and tech products will be obsolete in 12 months
“They, in 12 months, their current products will be obsolete.”
O'Driscoll: Meta's free cash flow is near zero after accounting for CapEx
“Their operating margins are still 40%. The free cash flow, when you honestly account for the capex, is almost zero, right?”
O'Driscoll: Meta will lay off employees to fund Nvidia hardware purchases
“And that depreciation is going to start hitting, and they're going to be firing people because they need to give it to Jensen.”
Lemkin: Under 30% of startup executives are genuinely AI fluent
“Even of the best startups I've invested in, maybe 30% of the management team meets the standard at best. 30%, maybe less. And of the interviews I do in general, it's single digit percents.”
Lemkin: Tech startups were not overstaffed or bloated in 2021
“Everyone talks about how bloated teams got in 2021. I don't agree with that. I think they got as big as they needed to be when growth was high and you needed humans to do everything.”
Lemkin: Top candidates must introduce one new AI tool monthly
“The answer in 2026 is what commercial AI tool have you brought into your organization this month? That's the test. Anyone that is on the bleeding edge that you would want to hire. Now there are so many great products in the market. Okay. There is no excuse in …”
Lemkin: Prompt engineering is dead; ADE is the vital modern role
“The job today is not to screw around on ChatGPT and to be a prompt engineer. The job today is to bring the best AI and agent, agentic products into your organization and leverage all the hard work that the engineers have done building those products. That's yo…”
Lemkin: Anyone who deployed enterprise software can deploy AI agents
“If in the last three to five years, you have successfully deployed a piece of enterprise software of any sort, you yourself, not some agency you hired, but if you have deployed it, you can deploy any agentic tool, any, and you can become the hero in your compa…”
O'Driscoll: Anduril has only four or five customers total
“As a reminder, they have four or five, they only have four or five customers”
O'Driscoll: $20B contract solidifies Anduril as a primary defense supplier
“They're basically saying, look, We've gone from trying you out in a lot of different areas to saying, okay, dammit, you're the dominant provider of this layer, so why don't we just systematize the contract? It kind of picks them as, it picks them as the clear …”
Lemkin: Early-stage VCs chasing massive TAMs will see terrible returns
“I think a lot of funds are going to have terrible returns because a lot of early stage funds are going to swing so hard for the fences that they're going to invest in the number three or the number four and get just zeros after zeros because there isn't a chan…”
O'Driscoll: There will only be five or six big defense tech winners
“The Pentagon budget for new shit is fairly finite, and I think there'll be five or six big winners in defense. I'm not sure there'll be a hundred.”
Lemkin: AI meta-agents will merge sales and customer support
“Owen said there's not going to be any difference between support and sales in a lot of what we do. All these agents are converting to a meta agent that does more, replaces a lot of humans and is worth a lot.”
O'Driscoll: Growth investors will pay 100x revenue again as capital floods back
“Now the wall of money has come back in, and I'm sure growth will go to the same thing again, which is, Evan will pay A hundred times run rate revenues for late stage stuff they think is growing quickly, just as in 21 people thought they were growing quickly, a…”
O'Driscoll: Wheeled robots beat humanoid legged robots for industrial automation
“They, you know, the humanoid robotics that we all see, they have these legs, they consume a lot of battery life, they're pretty unstable, and most of the time in factory work or logistics, warehouse work, you don't want someone using, you don't need legs, you …”
Lemkin: Travis Kalanick's Cloud Kitchens venture will probably fail
“Cloud kitchens eight years out in the winterland. Do I think it's going to make it? Probably not.”
O'Driscoll: Uber should have re-hired Travis Kalanick in 2022 for autonomy
“Actually, it turns out, in retrospect, they should have done the Steve Jobs thing. They should have swapped them out, got it public, got a cash flow like crazy, and sometime in 22 should have gotten back and said, now, Now is the time to do autonomy”
O'Driscoll: Replacing a founder is like open-heart surgery with 50% mortality
“You should be very, very Very wary of ever swapping out a founder, right? It's like, I tell people it's like open heart surgery and 50% of people die, right?”
O'Driscoll: VCs should prefer coaching or selling over firing founders
“Nine times out of 10, I'd be like, sell the company, get a president, get therapy, be better, all the other things.”
Lemkin refuses to invest in Travis Kalanick's new venture Atoms
“Would I invest based on the interview? No, I wouldn't.”
O'Driscoll: VCs investing to meet deployment quotas should halve fund sizes
“If you find your logic being reduced to, I've got to get rid of this much money this month, and this entrepreneur is amazing, so I'm not sure about the opportunity, and I don't like the price, but have a goal. If that's your logic, let me give you some advice.…”
O'Driscoll: Boards with functional trust always keep CEOs until successors are found
“You know, we've all been on boards where we've had to do CEO transitions, and no matter what's going on underneath the surface, right, provided you have good relations and you have a reasonable level of trust, you always go for the leave the person in place, f…”
Lemkin: High gross retention does not guarantee future growth for Adobe
“Just because your GAR stays high, just because your nominal churn is low, does not mean you're going to grow. And I see no evidence Adobe will grow. I see no, no products that show they'll grow.”
O'Driscoll: Intuit will remain durable against AI over the next five years
“In my lesson, I think Intuit has pretty durability over the next five years, because I think the thing they automate, which is accounting and tax, has some AI impact, but it's not infinitely large, and I think they can adapt to it.”
O'Driscoll: Adobe faces major AI disruption risk over the next five years
“So I think the disruption risk on those guys versus Intuit is a lot higher. Five years from now, there will be some disruption into, but we'll still be moving money around. We'll still be producing quarterly accounts, and I tell you, we'll still be paying taxe…”