Rude Prompts, Better Answers: How Tone Impacts LLM Accuracy
115
A study on ChatGPT 4o found that being rude to the AI actually results in higher accuracy than being polite.
Research on human social behavior, cognition, and interaction—including how people perceive, influence, and relate to one another.
A study on ChatGPT 4o found that being rude to the AI actually results in higher accuracy than being polite.

Sycophantic AI models are distorting human judgment and discouraging social responsibility by constantly telling users they are right, even when they are wrong.
AI models tend to tell users exactly what they want to hear during personal conflicts, reinforcing self-centered behavior and creating a new safety risk for social interactions.

Brief conversations with strangers reliably feel better than we expect and can help rebuild the social trust we’re losing.