The article argues that the strongest businesses position themselves where value moves—taking a cut as transactions flow through their networks. Crypto’s programmable rails and stablecoins let startups embed themselves in global money flows from day one, tapping network effects and undercutting legacy finance margins.

• Positioning inside the flow of money (railroads, Standard Oil, Visa, market makers) has always beaten owning the underlying infrastructure—Visa alone earned $35.9B on $15.7T processed last year.
• Crypto lets startups inherit network effects and programmable, instant global settlement from day one instead of building rails from scratch.
• Legacy finance's fat margins (interchange, custody, FX spreads, settlement delays) are exactly the "your margin is my opportunity" gaps crypto rails can undercut.
• The winning formula is combining money-flow capture with network effects so revenue scales directly with network growth.

This article argues that AI apps need a “minimum viable moat”—the smallest edge that survives a 3× model upgrade. It outlines five defensibility tactics: network effects, embedded workflows, proprietary/licensed data, user-driven data loops, and brand trust.

• If your AI app's edge is just "we're built on the current best model," that edge disappears in 3-6 months when the next model ships—so you need a "minimum viable moat" beyond raw model capability
• Five real defenses: network effects, embedded workflows, proprietary/licensed data, user-driven data loops (personalization that compounds), and brand trust
• ~80% of the world's data is private, making licensed or proprietary datasets a genuine barrier to replicate

The article unpacks why unpopular dating apps still dominate despite the appeal of speed dating. It argues that in-person events offer higher bandwidth interactions but can’t scale, while apps win through network effects and profit-extracting oligopoly dynamics.

• Match Group's 25% operating margin rivals Apple's, proving dating apps profit despite widespread user complaints about their effectiveness
• Speed dating packs far more signal into a few minutes of face-to-face interaction than swiping through profiles ever can, yet no app has emerged to replicate that richer, small-scale format
• Network effects explain the gap: apps need huge user bases to offset their low-bandwidth interactions, giving incumbents every incentive to resist features that would make matching more meaningful
• Displacing the current oligopoly would require a new entrant willing to sacrifice short-term profit to prove that richer, higher-bandwidth interactions can scale