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

2025-06-15

AI Summary

To address the proliferation of deepfakes and large-model-generated disinformation on social media, this paper proposes a three-tier anonymity governance framework grounded in a quantified reach score. Low-reach users retain pseudonymity; mid-reach influencers are linked to private legal identities; and high-reach sources undergo real-time, ML-assisted fact-checking. This framework achieves, for the first time, a scale-aware decoupling of anonymity rights, accountability, and content veracity. Its social feasibility is validated via spontaneous adoption mechanisms within Reddit communities. We develop an implementation pathway integrating a quantitative diffusion model, a lightweight ML-based verification system, and cross-jurisdictional compliance adaptations (U.S., EU, U.K.). Empirical evaluation demonstrates that the framework significantly curbs disinformation spread while preserving everyday user privacy—establishing a deployable, technology-institution co-design paradigm for platform governance.

Technology Category

• Application Domains: Misinformation & Fake News
• Natural Language Processing: Fact-Checking / Misinformation Detection (NLP Focus)
• Machine Learning: Privacy

Application Category

• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• User Modeling, Personalization and Recommendation: Accountability, Transparency, and Ethics for personalization
• Responsible Web: Mitigating misinformation and disinformation

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 $ extit{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.

Problem

• Counter deepfakes and LLM-driven misinformation on social media
• Implement tiered anonymity based on user reach score
• Ensure accountability while preserving privacy for small accounts

Innovation

• Three-tier anonymity framework based on reach score
• ML-assisted fact-checking for high-reach accounts
• Regulatory pathway adapting existing jurisprudence