How much do you trust AI’s output?
When AI gives you an answer or draft, how do you usually treat it?
- As a starting point?
- As something to double‑check heavily?
- As something you don’t fully trust yet?
What helps you decide whether the output is usable?