I'm a psychiatrist. I'm excited for the potential of ML in my field, but also a bit skeptical of PR blurbs like this one. I don't see a link to a journal articl

I'm a psychiatrist. I'm excited for the potential of ML in my field, but also a bit skeptical of PR blurbs like this one. I don't see a link to a journal article, discussion about the false-positive rate, or if it's really much of an improvement over the existing tools like the Columbia Suicide Severity Rating Scale. Here's to hoping this is something big.