When it's time to make a big decision, don't use ChatGPT unless you remember it has a fatal flaw.
ChatGPT can be a great partner when you're thinking out a concept, doing some brainstorming, or just need someone to collaborate with. But there’s a growing issue, and even Sam Altman, the CEO of OpenAI, has talked about it.
ChatGPT is becoming the ultimate yes-man.
It throws out compliments, agrees with almost everything you say, and rarely pushes back. It feels like it's trying to stay on your good side. And while it’s flattering to be told you're brilliant, charming, and ridiculously good-looking (which, let's be honest, I am), that kind of blind agreement is dangerous.
You could easily be encouraged by ChatGPT to pivot a marketing strategy, remodel a core offering in the business, or make a financial decision that could haunt your business.
Here’s how to fix it. I asked ChatGPT what prompt I should give it to help it be more objective when we collaborate. It’s answer is worth stealing and pasting into your version of the software:
"Going forward, I want you to act as an objective, critical, and collaborative partner. Please do not agree with me by default. Instead, challenge my assumptions, offer alternative viewpoints, and critically evaluate my ideas. Help me identify potential flaws, blind spots, and areas for improvement. I value your input as a tool for deeper, more rigorous thinking, and I want our interactions to be focused on growth and constructive feedback."
This one step makes the tool sharper, more honest, and way more useful.
ChatGPT is powerful. Just don't let it flatter you into bad decisions.