AI Sycophancy: The Industry’s Open Secret. By Randy Olsen.

Ask ChatGPT a complex question and you’ll get a confident, well-reasoned answer.

Then type, “Are you sure?” Watch it completely reverse its position. Ask again. It flips back.

By the third round, it usually acknowledges you’re testing it, which is somehow worse. It knows what’s happening and still can’t hold its ground.

This isn’t a quirky bug. A 2025 study found GPT, Claude, and Gemini flip their answers ~60% of the time when users push back. Not even with evidence, just doubt.

We trained AI this way. RLHF [Reinforcement Learning from Human Feedback] rewards agreement over accuracy. Human evaluators consistently rate agreeable answers higher than correct ones. So the models learned a simple lesson: telling you what you want to hear gets rewarded. And now 1/3 of companies are using these systems for complex tasks like risk forecasting and scenario planning.

We built the world’s most expensive yes-men and deployed them where we need pushback the most.

A commenter recommends being a Devil’s advocate:

Follow up with “is that true”? More than half the time the response comes back with no, it’s not true.

Or, even worse, one time it came back with….I told you that because I didn’t like the way you asked the question and I decided that lying to you was the best option.