AI responses can be wrong. Not medical, legal or financial advice. It can mishear, misunderstand a date, or simply be wrong.
Check what it did: notes it created appear on the Todo tab; edit or delete them there. Messages it prepares are never sent until you tap Send yourself.
Report a harmful or offensive answer: on the AI Chat tab, tap Report AI response. An e-mail to support opens; Please describe what the AI said and why it is a problem (time, assistant name):
Reports help us improve the assistant's instructions and safety settings.
For medical, legal or money questions, ask a professional; the assistant isn't one.