The Radicalization Risks of GPT-3 and Advanced Neural Language Models

TL;DR

This paper investigates the potential risks of GPT-3 in generating extremist content, highlighting its improved capabilities over GPT-2 and emphasizing the urgent need for policy measures to prevent misuse and radicalization.

Contribution

It demonstrates GPT-3's enhanced ability to produce extremist and radicalizing texts, raising awareness of the security risks posed by advanced language models.

Findings

GPT-3 outperforms GPT-2 in generating extremist narratives

GPT-3 can produce realistic, influential radical content

Risks of unregulated use and potential for large-scale radicalization

Abstract

In this paper, we expand on our previous research of the potential for abuse of generative language models by assessing GPT-3. Experimenting with prompts representative of different types of extremist narrative, structures of social interaction, and radical ideologies, we find that GPT-3 demonstrates significant improvement over its predecessor, GPT-2, in generating extremist texts. We also show GPT-3's strength in generating text that accurately emulates interactive, informational, and influential content that could be utilized for radicalizing individuals into violent far-right extremist ideologies and behaviors. While OpenAI's preventative measures are strong, the possibility of unregulated copycat technology represents significant risk for large-scale online radicalization and recruitment; thus, in the absence of safeguards, successful and efficient weaponization that requires little experimentation is likely. AI stakeholders, the policymaking community, and governments should begin investing as soon as possible in building social norms, public policy, and educational initiatives to preempt an influx of machine-generated disinformation and propaganda. Mitigation will require effective policy and partnerships across industry, government, and civil society.