Generative Persuasion

Algorithmic Propaganda

Generative Persuasion is an interactive installation and application that reveals how effective Large Language Models (LLMs) can be at quickly generating inauthentic content that can be shared online to sow division and confusion.

The project highlights the cognitive dimension of security that can be exploited through social engineering and is based on techniques that have been used by known influence campaigns conducted by the military, intelligence agencies, and corporations. Participants can micro-target groups based on their liberal or conservative bias and their OCEAN personality profile by algorithmically generating persuasive conspiracy theories, political ads and social media posts.

More content coming soon...

This project is currently in development.