Mm-hmm.

I think any reliance on AI without your own personal clinical review is automatically high risk.

There should never just be this delegation of a task, specifically a, a generative task, like report writing, without oversight.

And then, you know, related to that, I don't think we've got any information at this point in time that would give us confidence in using AI to do things like interpret assessments.

We cer- uh, certainly not there yet.

Uh, scoring tests, making risk calculuses or, you know, anything like risk assessment.

I think originally people were really hoping that AI would help us there, thinking that, well, computers aren't biased the same way humans are, so maybe this is our, our golden ticket to, to good risk assessment, and spoiler alert, it's not.

Do you feel this is a high risk or a low risk AI activity?