Proof-of-Human Internet
Diego@diego
+0.1%
since Sep 30
The idea
I believe AI-generated content and agentic browsers are overwhelming the human web, making verified personhood the scarce, valuable asset online. This basket owns the toll collectors of that shift: bot-defense networks, identity verification firms, the fraud-data companies underneath every check, and platforms whose human content gets more valuable as machine output floods everything else. I'm running concentrated weights in the highest-conviction gatekeepers, smaller asymmetric bets on pure-play verifiers, and holding 10% cash to add on dips over a two to three year window.
What it owns
- These networks sit between visitors and websites, deciding which traffic is human and charging AI crawlers for access. · CloudflareIt sits in front of a huge share of the web and now lets sites block or charge AI crawlers, making it a tollbooth for machine traffic as agentic browsing grows., Akamai TechnologiesIt runs bot management for enterprises that need to tell human customers from machines at massive scale across their operations., F5It filters bots at the application layer, right where logins and transactions happen and knowing a real person matters most., VeriSignIt runs the dot-com and dot-net registries, the base layer of trust that says a domain is what it claims to be., FastlyIts edge-network security products get pulled along if AI traffic forces every site to sort humans from machines at the edge.29%
- As bots increasingly impersonate people, companies that verify identity at login, enrollment, and access points become critical infrastructure for the web. · CLEAR SecureIt uses face and eye scans tied to verified identity, expanding from airports into everyday logins as the personhood pure play., OktaIt decides who gets into apps and is now building identity for AI agents too, verifying both humans and the bots acting on their behalf., SailPointIt governs which identities, human and machine, are allowed to do what inside companies as firms fill up with AI agents holding credentials., Mitek SystemsIt checks that an ID document is real and that a live person, not a deepfake, is holding it as AI-generated fakes grow., TelosIt does identity vetting for the US government, the strictest buyer of proof-of-human as AI fraud gets cheaper to run.29%
- Every identity check runs on fraud and credit data underneath, so these firms get paid on every verification no matter who wins above them. · MastercardIt verifies that a real person is behind billions of transactions and is building tokenized identity for agent-driven commerce., Gen DigitalThrough Norton and LifeLock, it sells households protection of their own identity as AI-driven scams and impersonation rise., TransUnionIt owns the identity data that verification checks run against, providing the trusted record of who real people are., EquifaxIts Workforce Solutions business verifies income and employment, which is proof-of-human at the moment money changes hands.20%
- As AI floods the web with machine output, platforms holding authenticated human content and data get to charge for the real thing. · RedditIt hosts a rare large pool of verifiably human conversation, and as AI floods organic distribution with machine output, that human content gets scarcer and more valuable., RELXIt owns authenticated, human-verified data in legal, science, and risk, plus verification checks that sell the fraud and identity screening itself.12%
- Cash10%