Fluency is not the same as truth
An answer can sound confident while containing invented facts, nonexistent references or incorrect reasoning. These errors are often called hallucinations. A confident tone is not a reliable measure of accuracy.
A model may also omit information, reproduce bias or answer the same question differently. Limitations vary by task and version. The greater the impact of an error, the more rigorous the checks and human oversight should be.
Use a draft message as a starting point. For a tax deadline, medical instruction or contractual obligation, confirm the source and consult the appropriate professional.
Responsibility for a decision does not disappear because a machine suggested it.
Try it yourself
Choose three tasks from your day. For each, write down the desired result, the data it needs and whether a simple rule would solve the problem. Explain one task for which you would not use AI.