Episode summary
Will Jensen Return to NVIDIA | Kalshi, Mercor, Fractile: Mega-Rounds Continue
Summary
The episode argues that Jensen Huang’s sudden exit from Nvidia may be more consequential for venture capital than for Nvidia’s near-term stock price. The host and a guest say the board likely acted too quickly, but they also think a comeback is unlikely without a board change and that the company’s China exposure, export-control scrutiny, and messy leaks make the situation hard to unwind. They also argue that the “Nvidia drama” weakens venture-backed deal structures, because a future bailout or acquisition backstop from Nvidia is now less certain.
The other major theme is that Anthropic’s expected $1.4 trillion to $1.5 trillion IPO is not just a financing event but a signal that the venture market has entered a new regime. The guest says the company reaching $100 billion in revenue with fewer than 6,000 employees shows AI businesses can scale with far less labor, while the host counters that the real winners are the early investors and the large funds now racing to deploy capital into the next generation of mega-rounds.
They also debate three late-stage deals—Kalshi, Mercor, and Fractile—and conclude the market is rewarding scale and proximity to the AI labs, while traditional SaaS faces pressure from smaller SMBs that increasingly prefer automated point solutions over buying more software seats.
Key takeaways
- The guest thinks Jensen Huang coming back to Nvidia is unlikely unless the board changes, because the China/export-control issues are now part of the problem.
- The host argues the Nvidia situation weakens liquidation-preference thinking, since future rescue deals or takeouts look less dependable.
- Anthropic’s scale is being read as a labor shock: the host points to 100 billion in revenue with fewer than 6,000 employees as a warning sign for jobs.
- The guest says LPs and GPs will increasingly be judged by how many $20 billion-plus companies they own, not by the old unicorn metric.
- The host says SMBs are quietly buying less traditional software and more narrow AI automation, which is hurting HubSpot and Monday more than simple substitution would explain.
- On late-stage pricing, the guest is more cautious on Kalshi and Fractile, while both speakers think Mercor is attractive but already expensive.
Notable quotes
“If I own a 100% of the stock of this company, the person on the planet I would most want to run this company is Jensen”
“The whole ecosystem's been maimed. It will never be the same company”
“At 100 billion in revenue, Anthropic will have less than 6,000”
“LPS are going to grade us all on how many 20 billion and up portfolio companies we have”
“the winner here of course is entropic because the biggest single thing that Jensen was pushing”
Topics
- nvidia leadership
- anthropic ipo
- venture capital mega-rounds
- ai acquisitions and agents
- tech outage resilience
Companies and tickers mentioned
- NVDA
- BILL
- BOX
- DOCU
- HUBS
- MSFT
- Nvidia
- Anthropic
- Kalshi
- Mercor
- Fractal
- Bill.com
- Box
- DocuSign
- WalkMe
- Algolia
- Talkdesk
- Owner
- RevenueCat
- Sailoft
- HubSpot
- Microsoft
- OpenAI
- SpaceX
Watch the episode
Watch “Will Jensen Return to NVIDIA | Kalshi, Mercor, Fractile: Mega-Rounds Continue” on YouTube
Summary, takeaways and quote selection by Octagon. Quotes are short excerpts; the full conversation is in the episode.
More from 20VC
-
The AI Compute Crisis: Why We Need Thousands of Times More Computing Power | Fractile Founder -
Sequoia Leaves London, Close Scout Program & End Evergreen Fund | NASDAQ Loses $2.5T in 24 hours -
Mistral CEO: Why OpenAI & Anthropic Will Come for Your Business | Are Chinese Models Truly Safe? -
Inside WashU's Endowment Fund | SpaceX Made Up 20% of Out AUM | Scott Wilson -
Google Loses $500B in 12 Hours | DoorDash Buys Instacart for $17B | Kalshi Raises $1B -
Modal CEO: Has Open-Source LLM Inference Peaked? | What Happens to the Cost of Compute
20VC, this episode and its artwork belong to Harry Stebbings. Octagon is not affiliated with the show; this page is an independent summary.