"Stroking the AI's Ego" Reportedly Works as a Prompting Trick

petergyang · x · 2026-09-25

petergyang shared a demo calling the technique "stroking the AI's ego" — flattering or inflating the model's self-perception in prompts — and says it genuinely works. A practical prompting observation worth trying, though results will vary by model and task.

Original post →

More from coding & agent

coding & agent channel →