Dr. John Torous

Because legally it says, we offer no psychiatric help, no medical help.

If you dare have a crisis, you've violated terms and conditions, you need to get off our service immediately.

And you go, whoa.

That's very strong language.

So sometimes if you're considering, again, I don't recommend these for treatment at all, but if you're considering using one, talking to one, look at what they claim the terms of service, because that's where you're going to figure out what the company really says in every single AI chatbot that I've seen in the world.

will disavow any clinical claims or medical liability in the terms of service.

Well, the question is, if they're not willing to kind of legally put their kind of words behind kind of risk, that's kind of all you need to know.

And...

It hopefully will get better, but that's where it's nice to have people like Holly who are saying, look, I want to find and celebrate the ones that can cross that boundary, but I also want to stop the ones that are not ready to be there too, so that we don't get confused.

especially when we talk about kind of the AI in these kind of therapy roles, you say, where did it learn all the information about therapy?

It kind of went to Reddit and social media and scraped it.

So it took all the information from people of lived experience.

Who is it trying to serve directly?

It's not trying to give therapy to therapists.

It may in part, but it's trying to then actually directly serve people

People.

So in essence, it's taking all the information from people.

It's geared towards people.

So it's only logical that the people it's designed for are part of the design.

And the people that you could argue it took or stole the information from are part of the design too.