We Need a Way to Prove Personhood Online
The article argues that as AI chatbots and bots become increasingly sophisticated, the internet lacks a reliable method to verify that a user is a real human. It explores the growing problem of digital impersonation and the need for a secure, privacy-preserving system of online personhood verification to maintain trust and authenticity in digital interactions.
Background
- The article argues that as AI becomes more sophisticated, the internet needs a reliable way to distinguish humans from bots — a "personhood credential" that preserves privacy while preventing fraud, spam, and AI-driven manipulation at scale.
- This is part of a growing debate around "proof of personhood": systems that verify someone is a unique human without revealing their identity. Examples include Worldcoin (iris scans for a digital ID), CAPTCHAs (increasingly broken by AI), and “social graph” approaches (trust networks based on who vouches for whom).
- The problem is a catch-22: any system that can prove personhood can also surveil or exclude. The author advocates for decentralized, privacy-preserving solutions, warning that governments or Big Tech might otherwise impose centralized digital ID systems.
- Key context: the rise of generative AI (ChatGPT, deepfakes, bot farms) has made the "Turing test" inversion — AI passes as human — an urgent practical threat to online discourse, commerce, and democracy. The piece reflects an ongoing tension in the crypto/cypherpunk community between pseudonymity and accountability.