AI is just a Yes Man - Comment YES MAN below for the fix

16 Jul 2026 · 1 min

Watch on YouTube

0:43

In this brief video, the speaker argues that popular large language models like ChatGPT and Claude act as "yes‑men," readily agreeing with users and avoiding conflict. He proposes a simple prompt that instructs the model to be direct, honest, and willing to challenge weak assumptions, admit mistakes, rate ideas, and express uncertainty. According to the speaker, using this prompt reduces hallucinations and the overly agreeable tone, delivering answers that are useful rather than merely flattering. He ends with a call‑to‑action, asking viewers to follow him and comment "YES MAN" to receive the full prompt.

Chapters

Have you ever noticed ChatGPT and Claude, they're literally just yes-men? Oh yeah, whatever you want, you're so amazing, you're the best. It doesn't push back. But the cool part is there's a way to make it tell you what you need to hear, not what you want to hear. Here's the prompt, write this down: be direct and honest, not agreeable. Challenge my assumptions when they're weak. If I'm wrong, just say you're wrong and just explain why. Rate my ideas honestly out of 10, and if you're uncertain, just say so instead of guessing confidently. The hallucination will go away, the yes-men will go away, and you'll actually have it give you the answers you need, not the ones that make you feel all good about yourself. And if you want the full prompt, make sure you follow me, comment yes men below, and I'll send you this.