AI chatbots sometimes guess when they don’t know an answer. I use a simple “cupcake prompt” to spot when an AI might be hallucinating — and it works surprisingly well.
Some results have been hidden because they may be inaccessible to you
Show inaccessible results