Governments Should Mandate Tiered Anonymity on Social-Media Platforms to Counter Deepfakes and LLM-Driven Mass Misinformation

View on arXiv ← Back to list

Authors: David Khachaturov, Roxanne Schnyder, Robert Mullins

Published: 2025-06-15 11:18:10+00:00

AI Summary

This position paper proposes a three-tiered anonymity framework for social media platforms to combat deepfakes and LLM-driven misinformation, where identity verification and fact-checking requirements increase with a user's reach score, balancing free speech with accountability.

Abstract

This position paper argues that governments should mandate a three-tier anonymity framework on social-media platforms as a reactionary measure prompted by the ease-of-production of deepfakes and large-language-model-driven misinformation. The tiers are determined by a given user's $textit{reach score}$: Tier 1 permits full pseudonymity for smaller accounts, preserving everyday privacy; Tier 2 requires private legal-identity linkage for accounts with some influence, reinstating real-world accountability at moderate reach; Tier 3 would require per-post, independent, ML-assisted fact-checking, review for accounts that would traditionally be classed as sources-of-mass-information. An analysis of Reddit shows volunteer moderators converge on comparable gates as audience size increases -- karma thresholds, approval queues, and identity proofs -- demonstrating operational feasibility and social legitimacy. Acknowledging that existing engagement incentives deter voluntary adoption, we outline a regulatory pathway that adapts existing US jurisprudence and recent EU-UK safety statutes to embed reach-proportional identity checks into existing platform tooling, thereby curbing large-scale misinformation while preserving everyday privacy.

Key findings

The analysis of Reddit data shows that volunteer moderators organically implement similar tiered governance structures as audience size increases, supporting the feasibility of the proposed framework. The authors also demonstrate that existing legal frameworks in the EU, UK, and US provide a potential path for implementing the proposed regulations.

Approach

The authors propose a tiered anonymity system based on a user's reach score. Lower-reach accounts maintain full pseudonymity, while higher-reach accounts undergo verification and, at the highest tier, independent fact-checking before content amplification. This approach is supported by observing similar self-regulating trends in Reddit's community moderation.

Datasets

Reddit data is analyzed to demonstrate the feasibility and social acceptance of a tiered anonymity system.

Model(s)

The paper discusses the use of ML-assisted fact-checking in Tier 3, but does not specify the models used.

Author countries

United Kingdom

← Previous