What AI cannot do
AI systems produce outputs that are likely, not outputs that are guaranteed true. A confident tone tells you nothing about accuracy.
They also lack common sense about situations outside their data, they do not know what happened after they were trained unless given new information, and they cannot take responsibility for a decision.
None of this makes AI useless. It means the person using it stays in charge: setting the goal, checking the result and owning the outcome.
An AI answer sounds very confident. What does that tell you about its accuracy?
Choose the answer you think is right.