3 AI automations I trust—and one I never will

After using and implementing AI for the past few years, I've noticed my trust falls into four buckets.

✅ Summarizing meetings

✅ Drafting emails

✅ Finding information across my notes and documents

❌ Sending messages to other people without my approval

My reasoning is simple: the first three can waste my time if they're wrong.

The fourth can damage someone else's trust in me.

I'm curious if others think about AI autonomy the same way, or if you've become comfortable letting AI communicate on your behalf.