AI — what do we want it to do?

AI — what do we want it to do?

I had a discussion with my dentist about AI and what it is doing for his discipline. In the medical field they have to take a lot of notes and they are hoping to use AI to listen and record the notes. Then the practitioner would review and edit them. This would save them a lot of administration time that could help especially General Practitioners that do not have an assistant and have to do both the note taking and interacting with the patient (dentists normally have an assistant to help).

My thinking did go off and I wonder if a practitioner could become complacent and not read/edit the notes. So there needs to be something in the process to say that they have read or edited the notes.

I believe the questions have not changed. When we start building anything the same questions come up

• Why are we building this? What is its intended purpose?
• Who is it for? What value does it add?

A focus on these questions are important.

I also feel it’s important more than ever to get in the room where they are talking about AI to ask the critical questions early.

The negative impact of these AI apps can be incredibly hard to digest. Examples are google apologizing for its racist app or the parents of a 16 year old sue Open AI.

We need to ask the questions not when the thing is built but at the start, through design, build and monitoring.

Ask “what is the worst that could happen?”