0
I’ve tried asking the same question in a few AI assistants and received answers that differ in emphasis, detail, and sometimes even the stated facts. The prompt was short and straightforward, so I didn’t expect such a wide spread.
How much of that comes from differences in training data, live web access, or the way each model is instructed to respond? When comparing answers, what should I check before treating one assistant’s response as more reliable than another’s?
