manuscript

When Large Language Models are more persuasive than incentivized humans
Large Language Models (LLMs) have been shown to be highly persuasive, but when and why they outperform humans is still an open …