The Stablecoin Future, Milei's Memecoin, DOGE for the DoD, Grok 3, Why Stripe Stays Private

Watch on YouTube ↗  |  February 22, 2025 at 00:22  |  1:45:41  |  All-In Podcast
Speakers
Chamath Palihapitiya — CEO, Social Capital
Patrick Collison — Co-founder and CEO, Stripe
John Collison — Co-founder and President, Stripe
David Friedberg — CEO, The Production Board
Jason Calacanis — Angel Investor / Founder, LAUNCH

Summary

All-In hosts Jason Calacanis, Chamath Palihapitiya and David Friedberg interview Stripe co-founders John and Patrick Collison. The first half covers Stripe's scale, its move into stablecoins through the Bridge acquisition, and why the Visa/Mastercard interchange economics are harder to displace than the rent-extraction narrative suggests, followed by Jamie Dimon's remote-work rant and corporate bloat, and Pete Hegseth's request for 8% annual Pentagon budget cuts. The panel then dissects the collapse of Javier Milei's LIBRA memecoin and rejects memecoins as unproductive gambling. The back half covers the Arc Institute's Evo 2 DNA foundation model, Chamath's takeaways from xAI's Colossus build and Grok 3, an asteroid impact update, and why Stripe has stayed private.

  • Stripe processes over $1 trillion a year, about 1% of global GDP, is profitable on a GAAP basis and has expanded into lending, issuing, treasury and billing.
  • The Collisons argue stablecoins are the first large new payments use case, driven by cross-border flows and emerging-market consumer demand for dollar balances.
  • Patrick says most interchange flows back to issuing banks and consumer rewards, so displacing Visa and Mastercard is a set of trade-offs rather than pure rent capture.
  • Fraud, not transaction cost, is the bigger economic drag on merchants, and Stripe positions itself as a reputation network across the internet economy.
  • Jamie Dimon's remote-work and bureaucracy rant prompts a debate on corporate bloat; Chamath blames off-the-shelf enterprise software for rigid roles and headcount.
  • Proposed 8% annual defense cuts trigger discussion of multipolarity, cheap drones versus carriers, and chronic defense-procurement inefficiency.
  • Milei's LIBRA collapse leads the panel to dismiss memecoins as unproductive gambling with a pump-and-rug dynamic.
  • Chamath says xAI's Colossus proved pre-training still scales, making the AI capex wave look justified and turning him bullish on Nvidia.
Ideas
John Collison Co-founder and President, Stripe 6:48
Stablecoins finally work as payment rails.
John argues stablecoins are finally happening because the technology is now good enough. Stripe followed crypto since the Bitcoin white paper and tried to make Bitcoin work as a payment method, which failed, whereas stablecoins on an Ethereum L2 or Solana now settle fast and cheaply. Stripe bought Bridge, the stripe of stablecoins, late last year, and he points to live usage - corporate treasury management, offering US dollar services to people all around the world, paying contractors in places like the Philippines where bank transfers are slow and expensive. He calls stablecoins the first really big payments use case.
John Collison Co-founder and President, Stripe 7:09
Bitcoin is gold substitute, not payments.
From Stripe's failed attempt to make Bitcoin a payment method, John concludes Bitcoin does not work as a transaction rail: transfers are slow, transactions are expensive, and you never know exactly how much you will receive because it is not denominated in dollars. But he says it is good as a store of value, effectively a gold substitute. The practical implication is that payments demand migrates to dollar stablecoins while Bitcoin's role stays monetary rather than transactional.
Patrick Collison Co-founder and CEO, Stripe 8:54
Card networks survive; stablecoins become underlying rail.
Asked when the Visa/Mastercard duopoly can finally be dismantled, Patrick pushes back on the rent-extraction framing: most of the interchange charged to merchants flows straight back to the issuing banks and then on to consumers in the form of credit and card rewards, and card programs are not actually big profit pools for most major banks. So any substitute is a set of trade-offs - fewer rewards, weaker consumer protections, less extended consumer credit - not a free win. He also expects stablecoins to become the common settlement rail underneath rather than to supplant the consumer-facing networks, which he thinks will be rebuilt on top of and substantially leverage stablecoins.
Patrick Collison Co-founder and CEO, Stripe 9:35
Emerging-market dollar demand drives stablecoin growth.
Patrick says the big stablecoin use case taking off right now is cross-border and outside the US: the vast majority of people in the world are subject to a weaker, more inflationary currency - he cites the naira devaluing by a factor of three or four in a couple of years - and consumers in those countries increasingly want to hold dollar balances, a use case that is really exploding. He frames it as a retail version of the eurodollar system of the 1970s and 1980s, where the minimum transaction was about a million dollars, while a consumer in Ecuador can now hold a one-dollar balance. He thinks this also deepens the dollar's status as the world's reserve currency.
David Friedberg CEO, The Production Board 46:56
Memecoins are unproductive gambling; avoid them.
Friedberg says he does not like memecoins and does not think they are productive or that they help rebuild the financial system: a bunch of people put money in and lose it while a few people make a lot. He treats them as effectively a digital collectibles business, no different from trading cards, but amplified perhaps a thousandfold because the friction, manual handling and shipping of physical collectibles is gone and the social feedback loop plays out in real time, which drives values up quickly and means buyers can lose far more than they otherwise could. He classes them as an entertainment and gambling mechanism and says people can do it if they want but he thinks it is stupid.
Chamath Palihapitiya CEO, Social Capital 79:36
Colossus proves pre-training scaling; bullish Nvidia.
Chamath had long assumed the base models were asymptoting, so he was not convinced the money being spent on Nvidia GPUs would ever be productive capex. xAI's Colossus changed his mind: 100,000 GPUs stood up in 122 days and heading to 200,000, and the resulting jump in Grok 3 quality showed there are still valuable gains left in pre-training, so a larger cluster still buys a better model. He says he was sneakily surprised by that pre-training upside, that it is very pro Nvidia, and that if it is true then all this capex may be justified and buyers can keep buying a lot of hardware - so he is now bullish on Nvidia.
Up Next

This All-In Podcast video, published February 22, 2025, features John Collison, Patrick Collison, David Friedberg, Chamath Palihapitiya discussing USDC, BTC, V, MA, MEMECOINS, NVDA. 6 trade ideas extracted by AI with direction and confidence scoring.

Speakers: John Collison, Patrick Collison, David Friedberg, Chamath Palihapitiya  · Tickers: USDC, BTC, V, MA, MEMECOINS, NVDA