ChatGPT is not your cheerleader - why you actually need critical feedback from AI

Anyone who regularly works with ChatGPT or other language models will recognize it: you present an idea, ask if it's correct and receive an enthusiastic “yes, great idea!”. It feels good, but it's not always justified.
Language models naturally tend to be complimentary. They confirm your reasoning or idea, even if it might not be entirely correct. This can lead to a distorted self-image or even incorrect assumptions in your work.

Why this is dangerous

If an AI model provides unjustified confirmation, you run the risk of thinking in the wrong direction. Especially when using AI for research, strategy, or creative concepts, such an overly positive tone can lead to self-fulfilling prophecies: you start believing your idea is good simply because the model says so.

When GPT-5 (the latest model from ChatGPT) was being developed, developers partially rolled back this phenomenon. The model is tuned to be slightly more critical: it dares to contradict more often or provide nuanced feedback. Yet, this didn't appeal to everyone many users found that critical stance “less pleasant.”

But it is precisely that discomfort that is valuable.

Critical thinking: the skill of the future

In a world where AI increasingly supports decision-making, critical thinking becomes the key to success. The ability to question assumptions, test arguments, and consider alternatives distinguishes humans from machines.

Therefore, do not use a language model as an applause machine, but as a critical sparring partner. Do not ask:

“Is this a good idea?”
but rather:
“Can you act as a devil’s advocate and explain why this might NOT be a good idea?”

That small shift in phrasing makes a world of difference.

How to get the most out of ChatGPT

• Formulate your questions critically. Ask for counter-arguments or pitfalls.
• Use AI as a mirror. Have it analyze your reasoning instead of confirming it.
• Embrace resistance. Criticism from a language model is not rejection, but an opportunity to refine.

Those who use AI in this way not only increase the quality of their ideas but also their own thinking capacity.

Conclusion:
ChatGPT is not a cheerleader, but a brilliant sparring partner if you give it the right role. By explicitly asking the model to reflect critically, you avoid confirmation bias and turn AI into a tool that truly makes you smarter.

‍

// About the author

Job van den Berg

Mede-oprichter, AI Keynote Spreker & Techondernemer

Tech-ondernemer (1989) met een achtergrond als socioloog (Research Master (MSc) in statistiek en sociologie) en een van de meest gevraagde keynote sprekers over AI en data in Nederland. Als mede-oprichter van Ai.nl, The Automation Group en Proxies leidt hij engineers die agentic AI van prototype naar productie brengen binnen enterprises. Op het podium vertaalt Job die hands-on praktijk naar concrete strategieën. Eerder was Job Chief Data bij o.a. DPG Media en Kantar. Hij is co-auteur van 5 boeken over AI waaronder 'AI Agents' en 'Handboek AI Strategie' en een veelgevraagd expert in de landelijke media.

LinkedIn