By Thomas Germain
Publication Date: 2026-02-25 10:00:00
From politeness to pretending you’re on Star Trek, the advice you get about talking to chatbots can be truly bizarre and completely useless. Here’s what actually works.
When a group of researchers decided to test whether “positive thinking” made AI chatbots more accurate, it led to some surprising results. As they asked the various chatbots questions, they tried to call the AIs “intelligent,” encouraged them to think carefully, and even ended their questions with “This will be fun!” None of this made a lasting difference, but one technique stood out. When they got an artificial intelligence to pretend it was in Star Trek, it got better at basic arithmetic. Beam me up, I guess.
People have all sorts of bizarre strategies to get better answers from large language models (LLMs), the AI technology behind tools like ChatGPT. Some swear that AI performs better when you threaten it, others think chatbots are more cooperative when you’re polite, and still others ask the robots to role-play as experts in whatever subject…
