How every result is checked, with a human in the loop.
Human-in-the-loop AI means a person you choose approves what matters before it goes out. Here are the five questions people ask before they start, each answered in a sentence or two and linked to the detail.
Who builds it, and has he done it?
I do. For two years I have used AI I built in my own daily work, reading and checking long documents.
About meHow do I know the result is right?
Every result is checked against its source before you see it, and work that fails the check is done again.
How I check for invented rulesWhat happens when the AI is unsure?
It passes the case to a person instead of guessing. We agree in advance who handles which cases.
See it hand questions to a personWho owns it at the end?
Your team. After handover your people run it day to day.
What your team keepsCan I see it working?
Yes. Watch a four-minute recording of the assistant answering three questions, with the sources checked.
Watch the recordingFor your technical team, the full tests and set-up notes are in the blog