Aug 14, 2025 · 1h 34m · 20vc
Sam Altman's Masterplan or a Gift to Anthropic? Palantir & Shopify Crush Earnings · 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, Harry Stebbings hosts venture capitalists Jason Lemkin and Rory O'Driscoll to dissect the commercial realities of the generative AI boom, the historic re-acceleration of mature software giants, and the impending workforce restructuring driven by AI-powered operational efficiency.
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 15.1% of the talking time here. How this is scored →
Harry as informed peer 4.6 Guest teaching 5.0 Guest disagreement 4.5 Harry pushing back 4.1
speaking balance: gold is Harry, purple is the guest (3 minute bins)
Sharpest disagreement ▶ 1:13:14 Rory snaps at Harry's forced leaderboard summary
When Harry tries to frame Rory's comments as a definitive public stock leaderboard, Rory forcefully rejects the summary and reprimands the host for misrepresenting his words.
Hardest push from Harry ▶ 8:09 Harry challenges Rory on Anthropic's competitive positionHarry refuses to accept the premise that OpenAI's GPT-5 pricing puts pressure on Anthropic, asserting that Anthropic executives are laughing at OpenAI's underwhelming release.
Biggest teaching moment ▶ 49:05 Rory breaks down Scott McNeely's valuation lesson for PalantirRory educates Harry on the mathematical limits of extreme revenue multiples, citing Scott McNeely's Sun Microsystems quote to illustrate why a 120x revenue valuation requires perfection.
Harry holds his own ▶ 22:17 Harry connects founder media presence to market trajectoryHarry draws on his direct interview experience with top AI founders to argue that quiet teams like Mistral and Cohere are actively suffering in market perception compared to vocal leaders.
the scores for every segment, with the reasoning behind each
| Chapter | Topic | Harry as informed peer | Guest teaching | Guest disagreement | Harry pushing back | Why |
|---|---|---|---|---|---|---|
| Is GPT-5 Underwhelming or a Strategic Business Win? | 6 | 5 | 6 | 6 | Harry challenges the view that Anthropic is threatened by OpenAI's pricing, arguing Anthropic is laughing at OpenAI's underwhelming launch. Rory directly rejects Harry's framing, saying he does not buy that perspective at all. | |
| OpenAI's Grand Valuation, AGI Scepticism, and the Nvidia Safe Bet | 5 | 4 | 2 | 3 | Harry draws an analogy between AI picks-and-shovels and crypto/Bitcoin, pointing to Nvidia as the undisputed core asset regardless of model winner. Rory agrees with the thesis while self-deprecatingly reflecting on his own tendency to overthink. | |
| Perplexity's Audacious Bid for Google Chrome | 6 | 5 | 3 | 3 | Harry contrasts vocal AI leaders with quiet founders like those at Cohere and Mistral, noting public perception labels quiet ones as losers. Rory breaks down browser unit economics and gently teases Harry for labeling mainstream users slow laggards. | |
| The n8n Three-Billion-Dollar Funding Round and Growth Fund Dynamics | 5 | 6 | 4 | 4 | Harry posits Accel's high-priced n8n round was a retaliatory move to counter Index's dominance in Europe. Rory and Jason push back on evaluating deals purely for brand or defensive posture, walking through growth return math. | |
| Datadog Earnings Analysis, Public Markets, and AI Capital Spend | 4 | 6 | 4 | 4 | Harry asks if Datadog's massive revenue concentration from OpenAI is a gift or a curse. Both Rory and Jason educate him that having a massive customer in the AI economy is always an absolute win, dismissing Wall Street's second-order overthinking. | |
| Palantir and Shopify's Breathtaking Re-acceleration | 4 | 6 | 3 | 3 | Harry asks how to evaluate Palantir's seemingly astronomical valuation. Rory cites historical re-acceleration data and Scott McNeely's famous 10x revenue warning to explain why high multiples carry long-term risks despite Palantir's dominance. | |
| The Ruthless Era: Employee Reductions and the AI Employment Reckoning | 5 | 5 | 8 | 6 | Jason aggressively argues that headcount reductions are necessary and B2B leaders must be ruthless, citing Shopify's 30 percent cuts during 91 percent growth. Harry pushes back on whether creating fear among young workers is healthy, sparking an intense clash. | |
| Monday.com's Multiple Compression and the Public Market Leaderboard | 4 | 5 | 7 | 5 | Harry attempts to force Rory into publishing a ranked leaderboard of public SaaS stocks. Rory forcefully shuts Harry down for putting words in his mouth, calling out the host's framing directly. | |
| Late-Stage Venture Capital and Extreme Capital Concentration | 4 | 5 | 3 | 3 | Harry highlights Carta data showing record seed valuations despite fewer total deals. Rory explains how multi-billion-dollar late-stage rounds distort market statistics and reframe VC asset allocation. | |
| The Myth of the Solo Unicorn and the Reality of AI-Powered Lean Teams | 4 | 5 | 6 | 4 | Rory bluntly calls the hype around one-person billion-dollar startups an idiot comment in the extreme. Jason re-grounds the narrative by showing how lean teams of 20 to 30 people can leverage AI tools for ultra-efficient $20M+ operations. | |
| Kalshi Quick-Fire: Market Cap, Stripe IPO, and xAI-Apple Legal Battles | 4 | 3 | 3 | 4 | Harry runs a rapid-fire Kalshi prediction segment covering Palantir's market cap, Stripe's IPO timing, and xAI's potential lawsuit against Apple. Rory and Jason share swift, pragmatic takes on litigation and valuation ceilings. |
Statements from this episode (34)
O'Driscoll: Predicting Wall Street Stock Movements Is Utterly Unknowable
“My big aha is it's like dealing with a deranged madman trying to estimate what the street will do. I spend no time on it. Utterly unknowable.”
Lemkin: Shopify Cut Staff to 8,100 While Revenue Rose 91%
“From peak employee was 20, 22, 11,600 employees at Shopify. Since then, revenue has grown 91%. Pretty impressive for a company at eleven billion in revenue. And employees have gone down from 11,600 to 8100. Gone down while revenue's up 91%.”
Lemkin: Non-Ruthless B2B Founders Will Lose to Zuckerberg and Karp
“He's ruthless, Zuck's ruthless, Karp's ruthless, and if you think you're gonna win in B to B, if you're not ruthless, you're gonna lose.”
O'Driscoll: AI has entered a 10-to-15-year period of incremental progress
“And then you settle into 10 or 15 years of incrementally making it better until you plateau, right? And from a zero to one perspective, that's not as interesting. But we're now in the grind it out, make it better, build a business stage of life, which I think …”
O'Driscoll: ChatGPT generates 70% to 80% of OpenAI's revenue
“ChatGPT, which is, you know, 70, 80% of the revenue dollars when they shipped, they didn't even think it was a major release.”
O'Driscoll: OpenAI and Anthropic can become $500B-$1T companies without AGI
“I think both of these companies can be plus or minus, you know, half a trillion to trillion dollar companies.”
O'Driscoll: OpenAI Will Hit $1T-$2T Valuation via Subscriptions and Ads
“Looks like its market share is pretty hard to change at this stage. So you start multiplying. You know, you'll have a paid tier of 20 bucks times a lot of people. You'll at some point have an advertising tier, and plus or minus your Google with a subscription …”
Lemkin: Two portfolio companies use AI software engineer Devin
“Two of my portfolio companies use Devin, like their AI tool, and they love it for niche use cases, but we never talk about it.”
ScaleVP talked to n8n in 2022 but underestimated workflow AI potential
“We had talked to them in about 22 or 23, and you're right, it, you know, What we clearly underestimated was, you know, and kind of this whole, you know, workflow automation, it's a category pre AI.”
O'Driscoll: Winners in crowded software markets are founders who ship AI fastest
“Which one of them will get the prize? I tell you, it always comes down to the founder who gets it the most, gets it the quickest, and just puts everyone in the room and says, no one's leaving till we're shipping an LLM-enabled version of this, and we're going …”
Harry Stebbings: 20VC's $2M-$3M high-valuation 'Yolo' fund pool is up 7x
“Do you know, the funny thing is we have a like Yolo segment of our fund, which is like, yeah, which is a tiny pool of cash, whatever, like two million or three million, whatever. But it's basically, we have amazing access because of the shows and people will g…”
O'Driscoll: VCs make most of their returns in one year out of seven
“You always, I mean, there's, as I say, there's one year in seven where you make most of your money, and the one thing you don't want to find is you don't want to not have assets to sell when that year comes around.”
Lemkin: OpenAI Pays Datadog $240 Million Annually
“So apparently OpenAI pays Datadog two hundred and forty million a year.”
O'Driscoll: AI CapEx Drives Almost All Current GDP Growth
“I mean, you see the data, almost the entire GDP growth Is AI CapEx.”
Lemkin: Palantir Re-Accelerated from 12% Growth to 45% Growth
“It goes from 12% growth at about two billion revenue in twenty-twenty-three to almost 45% growth at four billion in ARR.”
O'Driscoll: Only 1 in 10 Software Companies Re-Accelerate Growth for Two Years
“I look across 20 years, about only one in three companies re-accelerate for one year, and about one in nine, one in 10 re-accelerates for two years.”
O'Driscoll: Palantir's 120x Revenue Multiple Is Unsustainable
“At the same time, 120 times revenues plus or minus is probably not sustainable.”
Lemkin: Shopify Generates $1.3 Million in Revenue Per Employee
“Shopify is at 1.3 million in revenue per employee now.”
Lemkin: Tech Companies Do Not Need Half Their Workforce
“You don't need half your company, and Palantir and Shopify are proving it. You don't need half your company. You literally don't need half the people working at your company. You don't need them today.”
Lemkin: Sales Reps Quit on Day One When AI Monitoring Is Introduced
“Every single time it's been brought in, someone on the sales team has quit the first day. Quit every single time they quit the first day, including on our little Sastra team, someone quit in the day we brought it in.”
Lemkin: Shopify will shrink to 7,000 employees while growing 200%
“Shopify is gonna be at 7000 to 200% bigger.”
Lemkin: Success Rate Placing Veteran B2B Executives Dropped to 10%
“Listen, until maybe 18 months ago, if any seasoned B to B executive that I knew needed a job, ok, that I knew reasonably well, I'd get him a job by an email. I could reach out to someone, not even in my portfolio, extended network founders I met that I knew me…”
Jason Lemkin: Shopify handles 13% of commerce and 90% platform share
“13% of all commerce already goes through Shopify, right? And it has essentially 90% of the platform market share.”
Harry Stebbings bought Monday.com stock after post-earnings price drop
“Dude, I bought the shit out of it. I'm going to be honest.”
O'Driscoll: Median SaaS multiples returned to historical 6x norms
“For 20 years, up until 20, probably 2019, the median SaaS multiple was six point three-ish, and the median growth rate was almost 30%. And now, the median multiple is closer to six again. The growth rate's 20, but the profitability's more, right?”
O'Driscoll: Large Late-Stage Venture Financings Will Persist
“So as these companies stay private for longer, there's going to continue to be this steady stream of fairly humongous later stage financings.”
O'Driscoll: VC Market Will Not Return to $100 Million Funding Caps
“Yeah, I think it's maybe not as pronounced as now, but it ain't going back to everything with A's, B's and C's and nothing be more than a hundred million dollar raise.”
Lemkin: AI Scale Prevents Venture Capital From Returning to Old Model
“We're not going back. The AI is so much bigger than cloud and so different. And unless the LP is cut off the valve it's, We're not going back to the old venture.”
O'Driscoll: The one-person billion-dollar startup is a metaphor, not reality
“If you do, if you're running a company with a billion dollars in revenue, you're probably making some money. You got to file the tax. Maybe you can outsource everything, but you end up just, yeah, someone's got to meet the accountant. Someone's got to review t…”
Lemkin: AI will enable 20-to-40-person teams to build billion-dollar companies
“If you allow for some outsourcing agencies resources that come and go, I think we will see a bunch of 20, 30, 40 person, billion dollar companies because a core of engineers, a bunch of AIs making, and people are going to be like, you know what, dude, not, I d…”
Lemkin: SaaStr Will Generate $20 Million in Revenue With Two Employees
“If we're lucky, Saster Inc. Will do twenty million this calendar year with two people. Two people, and depending on how you count them, like two, two, two employees.”
Lemkin: SaaStr's AI SDR Booked Three Six-Figure Meetings in Two Days
“As I said, our AI BDR, we deployed this week, it set up three meetings in the last two days for six figure sponsorships.”