Verifying Humanity In An AI World | The Brainstorm EP 128
Most important take away
Proof-of-unique-human is moving from a niche crypto concept to a real business need as AI bots now produce more online content than humans, creating security, trust, and fraud problems for platforms like Tinder, Reddit, Zoom, and banks. World ID (formerly Worldcoin) is positioning itself as the privacy-preserving identity layer, but faces strong competition from Apple, Google, and Meta native sign-ins. Separately, the hosts expect advertising to boom over the next 5–10 years as AI-driven feature parity forces businesses to differentiate through brand spending.
Chapter Summaries
- World ID / Proof of Human: AI bots outproduce humans online; World ID’s “Lift Off” event announced integrations with Tinder (verified human badge), Zoom (anti-deepfake identity check), Reddit, AWS, and Okta. 18M people verified across 160 countries; credentials used ~150M times.
- Skepticism & Implementation: Nick argues iris scanning is creepy and impractical; World ID adds a selfie-based onboarding layer and positions itself as a privacy-preserving ID layer that holds other credentials (KYC, gov ID).
- Competition from Big Tech: Apple, Google, Meta already offer authentication. Hosts argue a third-party, privacy-preserving, portable standard could win if businesses prime the pump, though default behavior (Sign in with Apple/Google) is hard to displace.
- Regulatory Headwinds: Banned or suspended in Brazil, Indonesia, Hong Kong, Spain, Kenya. Bank demand could shift regulator attitudes once the $200M deepfake-wire-fraud scenarios become common.
- Advertising in an AI Bot World: Ad impressions could 20x from bot traffic; performance advertising will dominate further, and brand advertising becomes more critical as AI-driven feature parity makes businesses easy to copy. Ad-to-GDP historically stable at ~2% but could trend toward the upper bound.
- Tesla Robotaxi Expansion: Expansion to Houston, Dallas, Phoenix (with new charging infra). Scaling likely constrained by per-city learning of curb pickup/drop-off logistics plus “hallucinated routing” bugs.
- AI Model Competition: Anthropic monetizing enterprise well, OpenAI investing more gigawatts into training than Anthropic, xAI reportedly behind but reshuffling teams. Models will keep scaling.
- Blue Origin: Second US company to reuse orbital rockets; second stage underperformed and put AST SpaceMobile payload into wrong orbit (insured).
Summary
Actionable Insights
- Proof of unique human is a secular trend. Every platform will need it. The winning implementation may not be World ID, but identity verification will be foundational infrastructure. Watch for standards that win adoption.
- Performance advertising will dominate even more. Ad dollars will continue shifting from brand to performance because tracking is precise even with bot explosion. Meta (more performance, better algo) is expected to surpass Alphabet in ad revenue.
- Brand advertising becomes critical as AI commoditizes features. In a world of “solo-penuer” businesses with copied feature sets and bot armies, brand marketing is the moat. The hosts suggest someone should start a social-native brand marketing agency for AI-era startups.
- Advertising spend as % of GDP (historically ~2%) likely trends up. Heightened AI-driven competition drives incremental spend. Ad spend is moving into advertising + GPUs, not labor.
- Tesla Robotaxi scaling is a watch item. Charging infra build-out in Phoenix suggests aggressive fleet scale-up. Cybercab volume production this quarter and next. New-hardware Model Y FSD is described as “99% perfect” but still has routing hallucinations.
Stocks / Investments Mentioned
- Tesla (TSLA): Robotaxi expansion to Houston, Dallas, Phoenix; Cybercab production ramping this quarter and next. Positive framing on scaling trajectory, near-perfect FSD on new hardware.
- Meta (META): Expected to surpass Alphabet in advertising dollars due to superior algorithms and conversion on Instagram. Positioned as beneficiary of performance ad shift.
- Alphabet / Google (GOOGL): Large ad incumbent (~50% of digital with Meta), but expected to lose share to Meta.
- Amazon (AMZN): Referenced as ~$70B ad business; part of digital ad oligopoly.
- Anthropic (private): Monetizing enterprise well via Claude.
- OpenAI (private): Investing more gigawatts into training than Anthropic in 2026.
- xAI (private): Grok underperforming in adoption/monetization; internal shakeups reported; could still catch up.
- Blue Origin (Amazon/Bezos-related): Successful reusable orbital rocket first stage; second stage failure put payload in wrong orbit.
- AST SpaceMobile (ASTS): Payload on failed Blue Origin second stage; insurance expected to cover loss; eventual deployment still expected.
- Tools for Humanity / World (WLD token): World ID product and associated token; business-side integrations (Tinder, Zoom, Reddit, AWS, Okta, Bruno Mars tour tickets) are the new pump primers versus prior crypto-FOMO model.
- Coca-Cola (KO): Mentioned illustratively as a brand advertiser.
Key Use Cases Driving World ID Adoption
- Tinder verified-human badge (Japan first, rolling to US)
- Zoom call identity verification (post-deepfake CFO wire fraud incidents)
- Reddit unique-humanhood for certain accounts
- AWS + Okta re-verification for stolen devices / corporate account recovery
- Bruno Mars tour: allocating tickets to verified humans to prevent scalper-bot dominance
- DoorDash-style account de-duplication
- Banks / KYC as layered credentials on top of World ID
Risks / Counterpoints
- Iris scanning has a creep factor; low actual usage (~8 verifications per user on avg).
- Regulatory bans in multiple countries.
- Competition from Apple/Google/Meta native auth with larger distribution.
- Business model depends on external pressure (businesses forcing users to verify) materializing faster than Big Tech alternatives lock in defaults.