Can AI agents accelerate the spread of new ideas and behaviors?
The new research paper "The amplifier effect of artificial agents in social contagion," authored by Feng, M., Hitz, E., Lazzaro, L., Tănase, R., Algesheimer, R. & Mariani, M. S., explores how AI-powered agents influence the spread of ideas and behaviors through social networks.
The study compares human decision-making with that of artificial agents powered by five large language models. Across two experimental settings, researchers find that AI agents are more willing than humans to adopt new behaviors, even when fewer others have already done so. Simulations show that introducing a small proportion of AI agents into social networks can significantly increase overall adoption. The findings suggest that the growing integration of AI into decision-making processes may have important implications for social influence and the diffusion of ideas, technologies, and behaviors across interconnected human–AI networks.
Read the full article:
Feng, M., Hitz, E., Lazzaro, L., Tănase, R., Algesheimer, R. & Mariani, M. S. The amplifier effect of artificial agents in social contagion. npj Complex (2026). https://doi.org/10.1038/s44260-026-00101-2