Redditor proposes AI regulation framework: text detection, face verification, open algorithms

techtotechbytechy · reddit · 2026-10-18

A detailed Reddit proposal for AI and social media regulation:

AI text detection: a universal, mandatory token-detection tool on all platforms, highlighting AI-generated (red) vs human-written (green) text, with separate labels for dictated/lightly edited/grammar-corrected content to protect human writers and expose bot campaigns.
Anti-impersonation face verification: mandatory live face scan on profile picture uploads, with biometric data deleted immediately after verification and legally barred from AI training or secondary use.
AI media segregation: dedicated sections or platforms for AI-generated audio/video/images, plus an open-source scoring mechanism evaluating AI slop and human slop separately.
Algorithmic regulation: government-mandated de-amplification of abuse and coordinated hate narratives, amplification of constructive discussion, and user-controlled, transparent algorithms — citing Meta's beta feature letting users steer feeds via topic descriptions as a step in the right direction.

The author also asks whether a UN tech department with AI-focused subdivisions is needed.

More from AGI Musings