Transcript

AI: The new frontier for mental health support?

0:01 The very best founders I know are brilliant at building systems. They connect teams, they remove bottlenecks, and they eliminate single points of failure. And yet When it comes to their own wealth. Most are running a disconnected stack. A tax accountant here and a state attorney there, a wealth manager who doesn't talk to either one of them.

0:22 Creative planning was built to fix exactly that. One integrated team of tax professionals, state planners, investment specialists, all coordinated by a dedicated wealth manager who sees your full financial picture and keeps every piece working together. Proactive tax efficiency, state strategy, investments all under one roof. Creative planning where wealth works together. Learn more at creative planning dot com slash masters of scale.

0:53 Hey folks, Jeff Berman here. I am thrilled to share some of the new names who will be joining us at this year's Masters of Scale summit. This may be our biggest stage yet. Reed Hastings, Meredith Whitaker, Van Jones, Amjad Masad, and more. Will be there with us October 20th through 22nd in San Francisco. If you're building something great, or you want to build something great, We want you there with us too.

1:20 Join us at masters of scale dot com slash apply twenty six. That's mastersofscom slash apply. Twenty six. Hey everyone, Bob here. Today we have a special episode spotlighting the relationship between AI and mental health.

1:43 Recently, OpenAI has faced several lawsuits, alleging that its chatbot contributed to suicides or mental health breakdowns. We're sharing a conversation that I had a few weeks before the suits hit with two on the ground experts, Brown University's Ellie Pavlik. director of a new instituted to exploring AI and mental health, and Saria Durabi of VC firm TMV, an early investor in mental health AI startups. Ellie and Saraya talk candidly about the pros and cons of applying AI to emotional well-being, from chatbot therapy to AI friendships and romance. Record live at the Innovation at Brown Showcase in Providence, Rhode Island. The conversation includes themes that are controversial and for some listeners potentially distressing, so please take care. I'm Bob Safian, and this is Rapid Response.

2:41 I'm Bob Safian live at Innovation It Brown Showcase and I'm here with Brown University's Ellie Pavlic. director of a new institute on AI and mental health based at Brown. And uh Saria Derabi, uh lead partner in VC firm TMV. Which has two hundred million dollars under management and is a backer of mental health well being startups, among many other things.

3:07 Ellie Saroya, thanks for joining us. Good to be here. Nice to see you, Bob. So Um, among the many fields that have been disrupted by generative AI. Mental health and the personal relationships we build with AI agents are among the most

3:24 Fraught? A recent study showed that that one of the major uses of chat GPT for users is mental health, which makes a lot of people uneasy. Ellie. I wanna start with you. The the the new institute that that you uh direct known as ARIA, which stands for

3:42 I have to read this to get this right. I always have to read it too. AI Research Institute on Interaction for AI Assistance. It's a consortium of experts from a bunch of universities backed by twenty million dollars in National Science Foundation funding. So what What is the goal of ARIA? What what are you hoping it delivers? Why is it here? You know, mental health is something that is very, um, I would say

4:07 I don't even know if it's polarizing. I think many people's first reaction is negative at the the concept of AI mental health. Um So as you can tell from the name, we didn't actually start as a group that was trying to work on mental health. We were we're a group of researchers who are interested in the biggest, hardest problems with current AI technologies. What are the hardest things that people are trying to apply AI to that we don't think the current technology is quite up for? And mental health came up and actually was originally kind of taken off our list of things that we wanted to work on because it is so scary to think about um if you get it wrong, like how how big the risks are. And then we came back to it exactly because of this. We basically realized that, you know, this is happening. Like people are already using it. There's companies that are like startups, um some of them d probably doing a great job, some of them not.

4:53 The truth is we actually have a hard time even being able to differentiate those right now. And then there are, you know, a ton of people just going to chat bots and using them as therapists. And so we're like the worst thing. that could happen is we don't actually have good scientific leadership around this. How do we decide what the technology can and can't do? How do we evaluate these kinds of things? How do we build it safely and in way that we can trust there's questions like this, there's a demand for answers. And the reality is most of them we just can't answer right now. Like they depend on an understanding of the AI that we don't yet have, an understanding of humans and mental health that we don't yet have, uh kind of a level of um discourse that society isn't up for. We don't have the vocabulary, we don't have the terms, like there's just a lot that we can't do yet um to make this happen the right way. So that's what Arya is kind of trying to provide this. You know.

5:39 Uh public sector academic kind of voice in to help lead this discussion. You're not waiting. for this data to come out or for the final, you know, whatever academia might say uh this consortium might say about it, you're already investing in companies that do this. I know you're an early stage investor in Slingshot AI, which delivers

6:03 mental health support via the app Ash. Is Ash the kind of service that You know, Ellie and her group should be wary about like ha w what what were you thinking about when you when you decided to make this investment? Well

6:20 Actually I'm not hearing that Ellie's weary. I think she's being really pragmatic and realistic. In broad brush strokes, just zooming back and talking about like the the sobering facts and the scale of this problem. one billion out of eight billion people struggle with some sort of mental health issue. Um, fewer than fifty percent of people seek out treatment, and then the people who do find the cost to be um prohibitive that recent study that you cited, it's probably the one from the Harvard Business Review, which came out in March of this year, which studied use cases of chat GPT, and their analysis showed that the number one, four, and seven out of ten use cases

6:58 Four. Foundational models broadly. are therapy or mental health related. I mean We're talking about something that touches half of the planet. If you're looking at investing with an ethical lens,

7:11 There's no greater TAM. than people who have a mental health. Disorder of some sort. We've known the Slingshot AI team, which is the largest foundational model for psychology, for over a decade. We've followed their careers, um, we think exceptionally highly of the advisory board and panel they put together. Uh But you know, I think what really led us down the rabbit hole of caring deeply enough about mental health and AI to frankly start a fund dedicated to it. And we did that um in December of last year, was really kind of going back to the fact that AI therapy is so stigmatized.

7:49 And people hear it and they immediately jump to the wrong conclusions. They jump to the hyperbolic examples of suicide. And yes, it's terrible. There have been Incidents of deep codecendence upon chat GPT or else otherwise, whereby young people in particular are susceptible to very scary things. Um And yet Those salacious headlines.

8:10 Don't represent the vast number of folks whom we think will be well serviced by these technologies. I mean, Ellie Tools are moving and changing so fast. Like w how do you think about what

8:25 you know, your research can do and sort of what the impact can be. Or You know, or is it as much about sort of creating guidelines that will help folks Like Soroy and the folks who work at Slingshot to

8:40 Navigate. So there's kind of a few, I would say like lanes in which we're thinking about things that will kind of converge and inform each other, but you can think about separately. Um So one is Things like guidelines, how to design systems and how to evaluate them. We're really thinking about like the development process for AI right now. Like what is the current process and what about that is not well suited for mental health? Like I would say, I was initially fundamentally skeptical. And one of the things that's made me more optimistic is the most optimistic in our team about the use of AI mental health are the people who will work most closely on mental health. So the the physician.

9:16 the um the people who have spent time working on mental health technology and like there's real potential here, right? Um the people who are most scared are the researchers who do basic science and have never thought about mental health. I also hear from students All the time I got an email earlier today with someone who's like, I was so excited you work on this because me and my friends have all been talking about what a great like um guide these chatbots have been. Right. So So there is real potential. But part of the problem with this development is it's not being developed for that use case, right? We're building these huge generalist systems and the same system that's gonna be churning out new chemistry compounds is also going to be. doing mental health support and is also going to be helping you cheat on your homework and is also going to be like formatting Excel formulas. Like

9:59 What, right? Like maybe, maybe like we're in an innovative like maybe that's true, but like we shouldn't take for granted that that's what this looks like. So I think one thing is just Not getting to much in the exploit phase too soon of being like look we have a recipe for building Big AI systems, and it's exciting, and therefore that's what we're doing. And we're just gonna run with this. Um, but actually opening the discussion and seeing what do we want AI to do, what about our system might produce that technology and what won't. And from that perspective, we're really thinking about these more participatory design kinds of processes, thinking about, yeah, what do we want the technology to look like? Let's not assume it's a chatbot. Maybe it is, maybe it's not like let's actually let's think about this because um the stakes are too high to just be like let's run with the first thing.

10:42 I think having companies trying to do stuff Is part of that, right? Like we can't just sit in a room and like think about it forever. Like we need to prototype stuff. There is a basic science development part, right? So there's um, like I said, these big journalist systems. The reality is we know so little about them. Like my own lab's work, and a lot of our work is kind of on understanding how language models work, like what's happening inside. Um, because we kind of stumbled upon this. So when people talk about making them safe or placing guardrails, what we're really doing is guessing and checking, right? We won't really know what's happening. And we do need Scientific leadership in conjunction with stuff being deployed like that needs to happen in real time now. And then we also need to be like, we're not like this is not the end of it. This is not the AI. And now we basically have this and we either have to make it work or scrap the whole thing. Right. Like we're kind of at the start of stuff. So there needs to be a discussion about like what else might AI look like. What other learning algorithms, what other architectures, what other models and

11:38 What other interfaces? Like there's so many options there. He he said this phrase like you know, we kind of stumbled on this, so one, four, and seven uses for for chat GPT. Like the You know, it's not what it was created for, and yet People Love it for that. It makes me think about Um, twenty years ago when everybody was freaking out about the fact that kids were in

11:59 Video games all day, and now because of that, we have Khan Academy and Duolingo. Fear mongering is good actually, because it creates uh a precedents for the guardrails that I think are absolutely necessary for us to safeguard our children from anything that could be disastrous. But at the same time, if we run in fear, we're just repeating history, and it's probably time to just embrace the snowball, which will become an avalanche in mere seconds. AI is going to be omnipresent everywhere. Um, everything that we see and touch will be in some way supercharged by AI. So if we're not understanding it uh to our deepest capabilities, then we're actually doing ourselves a great disservice. to this point of yeah, people are drawn to

12:37 AI for this particular loose use case. So on our team in ARIA, like we have a lot of computer scientists who build AI systems, but actually a lot of our teams do kind of developmental psychology core cognitive science, neuroscience, there are questions to say like Like Why, right? The whys and the how's. Like what what are people getting out of this? What need is it filling? I think this is a really important question to be asking soon. I think you're completely right. Fearmongering um has a positive role to play. You don't want to get too caught on it and you can point historically to examples of like people freaked out and it turned out okay. There's also cases like social media, like maybe people didn't freak out enough, right? Like and I would not say it turned out okay. People have to can agree to disagree and there's, you know, there's plus and minuses, um

13:22 But the point is like these aren't questions that really Like we are in a position that we can start asking questions. You can't do things perfectly, but you can start to you know, you can run studies. You can say what it like what is the process that's happening? What is it like when someone's talking to a chat bot, is it similar to talking to a human? What is missing there? Is this gonna be okay long term? What about young people who are doing this in core developmental stages? What about somebody who's in a state of acute psychological distress as opposed to as a general maintenance thing? What about somebody who's st struggling with substance abuse? Like these are all different questions. They're gonna have different answers Again, I think just I feel very strongly that the one LLM that just is One interface for everything is

14:00 I think a lot is unknown, but I would Bet that that's not gonna be The final thing that we're gonna want. Right. And I think we can do these things carefully if we're having conversations in the open. Um, and if we're having like healthy disagreement, like you have companies trying to move forward and people pushing back and regulators and a lot of people and part of the conversation, we can make this happen the right way. At the right pace.

14:22 Um we don't want fear of everything to cause break into what I think AI commonly is now, which is like the hypers and the naysayers, and this prevents any kind of reasonable middle ground progress because you're basically like with us or you're against us. And then that's um that's not productive. But I think we're at a a point right now where we can have like a really sober conversation and say, We have an opportunity here, but let's not um Let's not move too fast, but let's not let fear prevent us from doing anything.

14:48 You mentioned Soraya that You mentioned kids. I mean adults also are using these tools, right? I mean it's it's not it it's not just Kid. Well

14:58 In fact, it's not just adults, but it's predominantly men, and this is super interesting. So in addition to Slingshot, we're also investors in a company called Daylight Health, which allows nurses and and and folks who assist doctors, the ability to offer mental health services, which PCPs, primary care physicians normally don't have time to do. But we learn in our research and diligence for that particular investment that the vast majority of men feel too nervous to admit to anyone other than their primary care physician that they're dealing with symptoms of depression and anxiety. And the most beautiful part about AI therapy, if you use it well, or technologies like Daylight Health is that it provides a certain amount of accessibility and acceptance. I'm thinking about sort of the friendships and even romantic relationships that people have

15:50 started to develop with with AI agents, you know, when open AI released Geep G P T five, like there's this outcry because thousands of people lost their Boyfriends and girlfriends are, you know, um, they're Reddit groups with thousands of people on them about virtual romantic partners. Like, Is that Progress like is that opportunity?

16:11 For Allowing people to have connection that they crave and need, or or is it just scary? Again, I know I'm going to the extremes again, but I mean maybe. I heard someone in Silicon Valley Someone high profile say that Gun violence is completely out of control in America. And maybe what we need to do is get AI romantic companions. And I I mean that's that's an extreme example, right, of a particular solve for a terrible epidemic. But You know, we can't immediately write off the fact that AI relationships are bad because it's not what we're normally used to. I know that um

16:48 I mean, I love the movie Lars and the Real Girl because I think it that movie in particular addresses the the thought of a fake companion so beautifully and so eloquently in that, you know, we're all just out there seeking community and companionship. And whatever form it comes in should be socially acceptable. I agree. I think, you know I'm first and foremost a scientist, so it's like I don't There's not a huge value to us speculating on this, like, is it good, is it bad? I think there is an instinctive

17:15 like judgment that comes to it, like, oh, that's weird. But like Like this is, you know, again, this kind of classic, like something is new and different. The default because so much of the discourse around AI has kind of like a human. So we think of it as replacing human companionship, but a pretty powerful analogy that I think it's also maybe more like right now is journaling. And that's actually something that would be great if more people did this personal reflection, like actually externalizing your thought process. And I know some of the um the people who work on mental health within ARIA have been looking on these kind of AI assisted journals.

17:49 You know, a good therapist will do this, kind of just get you to think through your own life experiences to reflect. Practice gratitude, you know, think about your goals and ambitions, work through something that you might be having trouble voicing to other people. It's a really valuable process to go through. So we shouldn't just right away say the process of somebody Playing out a fantasy, working through an imaginary scenario, talking While being by themselves in a room is a bad thing. It could be a positive thing. Um, if it's replacing other things, it could be a negative thing. So I think there's again. Just questions to ask and research that needs to happen to say

18:24 that we should be aware of and start tracking it and hopefully we can Course correct if it's going the wrong direction. I wasn't quite sure how Ellie and Saroya were gonna respond to my questions about AI-based relationships. The idea makes so many people uncomfortable, but as they both say, if folks are using it, then we shouldn't just dismiss it. Still, we do need to understand it. So how do we judge whether AI is helping or hurting our cognitive state?

18:55 We'll talk about that. After the break. Stay with us. Hey listeners, Bob here. If you listen to Rapid Response on Masters of Scale, you may be missing half the show. Because every Friday we release a second rapid response exclusively in the Rapid Response feed. The guests and topics are just as compelling and timely from Ford CEO to NASA's administrator to the lessons from The Devil Wears Prada. It takes about

19:30 10 seconds to find, just search rapid response wherever you listen to podcasts and hit follow to make sure you never miss an episode. I hope to see you there. Humans will never be more intelligent than AI. There's gonna be two types of companies. Those who were great at AI and those that went out of business because they weren't. How do we build a future? That is human centered.

19:56 I'm Rana Al Chayubi. And on my podcast Pioneers of AI, we answer that question and so many more. As an AI scientist, entrepreneur, and investor, I know what it takes to build AI that works for everyone. Every week, I sit down with the pioneers shaping our future. And we take you behind the scenes of the AI that's transforming our lives. Find pioneers of AI wherever you tune in. Mm.

20:26 Before the break, we heard Brown's Ellie Pavlik and TMV Seroya Darabi talk about the rise of AI-based mental health treatment. Now we explore how AI might change us as humans, the challenges in effectively evaluating AI mental health tools, and more. Let's jump back in. Yesterday I was in an event. Where uh Siddhartha Mukherjee, the author of the Emperor Hall Maladies, was talking. He's got an AI company called Manus AI. Um and and he he was saying how Customer service representatives.

21:02 That are now like becoming AI agents. That like once you're done with your customer service. That agent may say to you, So what else is going on? You know, how else are you doing? And it's like he's saying it's gonna be like this your your customer service agent and your therapist are like gonna come together. And I don't know whether when you hear that like

21:23 Ellie's laughing. I don't know whether Saroya's thinking like Hmm, is that a market I can like but are we gonna need to train all of the agents that are everywhere? To be more understanding about You know, the mental health impacts because it's so alluring for us to have that. Conversation Oftentimes when people talk about AI therapy, some of the big headlines for what one might be concerned about include AI can't be empathetic because it doesn't have real lived experiences. Um

21:58 No, I I I'm not interested in these add-on experiences and and making everyone a bot or everything a bot. And also, um AI will figure out empathy soon enough, and some argue it already has. Because the data of speaking to milds of people. About millions of problems will ultimately result in some version of a lived experience.

22:21 And through that We can create this inference layer. of hopefully f ethics. And empathy that may be applied to other things. So maybe Not the customer experience guy, but maybe when you go to the grocery store, the grocers will be trained on how to speak to you more cordial.

22:38 And that might be rooted in data and yeah, perhaps that's really cool. Is there a way you research Differently. When you're I mean, you're looking at sort of the cognitive reactions we're having, right? Is there a way you research that differently because it's an AI, or or is it sort of We're humans.

22:57 AI's not changing the way our brains work. So like it's the same. First I want to touch on this this concept of empathy because I think this is a really good example of what I mean About there being like questions we want to answer right now and just the levels at which we don't have the scientific understanding to answer them.

23:17 We take for granted humans have empathy and lived experience and models don't. Um, the truth is we don't we don't have really good definitions of these things, right? What do we really mean by the a word like understand or empathize. Like when we talk about like it's it's important for humans to get empathy from another. Like Like be more specific. What do they actually need? Because some of these things we can imagine approximating with AI, some of these things we can't, which is the one that has the

23:44 Positive effect on that person's life. This is a really good opportunity to not pretend we know more than we do. You think about something like CBT. As a as a um Therapeutic device. You know, trials of AI assisted bots that are basically just reminding you to think through things, kind of reprogram your brain.

24:04 This doesn't depend on like a deep psychoanalytic, like, oh yeah, no, your mom was totally the one in the wrong there. Like that's not an important part and can be counterproductive, right? In other cases, we actually really feel like Okay. You can tell a human interaction you have where the person's Faking it. Versus not, and you sometimes filter out the friendships where you're like, oh, this person's saying things, but they don't mean them, right? And so there's some.

24:28 We need to figure out w how important these different things are, how much of this is part of how we're raised and programmed and and kind of innate and how much is something that's societally programmed and will drift over time as AI starts to fill a role and we start looking for other types of things. So much is on the table, so much is unknown. So I think it's just impossible to anticipate. Um uh anticipate what direction this is gonna go.

24:51 Um on your last question, like A I will not change how w I think that's just Definitely wrong. AI is absolutely going to change how we think. Um if nothing else that just knowing that. AI is out there as one of the things we might be interacting with will change our expectations. That's one of the most um Fascinating.

25:09 Whether good or bad from now looking forward is like how are we as individuals and as a society going to change as a result of having this in the mix? What I'm also hearing Ellie describe is something that we're seeing play out within um nine one one call centers. So AI has already Um

25:26 Basically infiltrated call center. So when somebody Dials nine one one. You might not know this, but initially you're being screened. Are are you experiencing a true emergency or non emergency? And now with the advent of AI, um, including companies I know well like hyper AI. You're able to immediately within seconds figure that out so that the true emergency calls are rooted to real humans and the non-emergency calls are redirected my cat's stuck in a tree and and a local firefighter might be able to go and help the cat fall out of the tree so that safely so that the person who is

26:00 Held up at gunpoint may actually find the floor. kind of emergency response they need. I think there are perfect parallels within SMEs, a serious mental illness. And for instance, your example of CBT, cognitive behavioral therapy, whereby people who just need on repeat the training to work through their day to day anxiety can get that access at scale so that we're redirecting the SMEs to people who actually can provide that high touch Empathetic um and bespoke concierge service. There's so much emotion around this topic.

26:34 How do we know Whether we're making progress. By far the hardest problem in AI is evaluation. There might be a small number of things. That are easy to evaluate, but all of the things that are really getting people excited about AI is gonna disrupt these sectors really depend on having

26:52 good evaluations of on things that have thus far not even necessarily been quantitative sciences. Um mental health is a really good example. Um education is another one that's a really good example. Um Any of this other kind of like having AI help with things like managerial work, any of this kind of classic white collar stuff, leading teams, coming up with ideas, doing science. And like And he will Anyone can pick their field and think about how they're evaluated.

27:18 in their own job or how their children are evaluated in school, and no one is satisfied with these evaluations. The truth is We all have good quantitative metrics for most things. We have a lot of proxies and corlets, and we know that they make mistakes. AI really depends on. Very specific, um success metrics, which is why we're seeing like progress in code. A is very good at writing code because it's pretty easy to evaluate if the code was correct. Um, and in other cases, we're just assuming that because it worked for code and because people who write code think that that's like the top of the hierarchy. Like if it can do code, surely it can do everything else.

27:51 Jokes on them. That was the easiest thing. Like by definition, this other stuff is by definition m harder. That's why we haven't figured out how to codify it yet, um, or how to quantify it yet. So Yeah, so for mental health, we're gonna have to define success. It's not gonna be we labeled a data set on like this was the correct response, this was the incorrect response, and have a leaderboard and see who gets to the top and do reinforcement learning against it. It's not gonna be that ARIA big focuses on this participatory AI. So it's definitely not computer scientists and AI researchers and companies who get to define this and then crank on the metric. It's gonna be a large, messy conversation. It needs to include people who think AI is absolutely the wrong way to approach mental health, people who are super gung ho, everyone in between. Everyone needs to be a little bit involved in this so that we come up with like

28:41 What does success look like? And we really don't know, like I think until we've had this really large conversation of what do we want to look like, we just can't really even speculate on what it's gonna be. Except to say that I will speculate, which is this is not gonna look like what current AI evaluation looks like, which is a leaderboard. When you think about AI's impact on Mental health.

29:00 Excitement, optimism. Potential. On one side Risk.

29:07 Fear. you know, sort of destructiveness potentially on the other side. Where where in that spectrum Do you feel like we we fall? I'm always a cautious optimist when it comes to technology. I don't think you could do What we do for a living and not be.

29:23 So Um I'm I'm somewhere just above The middle zone. And also

29:32 The coming wave has Crashed onto shore. Um, and if we look at everything through this dystopian lens, which is quite easy to do. We're missing the plot, I feel. And the plot is it's here.

29:45 It's happening. In this instance people are using Foundation models, Claude AI, Chat GPT to solve Everything from I'm nervous to go to work today to

29:56 I'm a doctor who needs an AI scribe to help me keep up with my patient load. And um Industry is capitalizing on it. Okay.

30:06 Um And society should also work with folks like Ellie who have the nuanced academic and and research on how we can tread water as carefully and safely as possible. I would say that I

30:22 Right now I am optimistic and there's like and it's'cause I think we're still in this narrow window. where we get to have a choice to make this happen the right way. I can imagine myself quickling that optimism if I feel like that window closes and like the main narrative I hear about AI is how fast things are moving. So I think right now, you know, it depends on the technology, it depends on people being sober minded, thinking about it. not being histrionic in positive or negative direction, that we're thinking about this. А за хол соціаліз овар проблем. I think we have this window still where we can like get it right.

30:54 I wanna think both of you, Ellie and Soraya, for being here and doing this. Thanks so much. I went into my conversation with Ellie and Saroya thinking that they might be at odds about AI's role in mental health. What I find telling is how aligned they are.

31:11 We can't ignore that people are using AI for emotional support. Yet we also can't ignore that most general AI tools were not built for this specific use. As Ellie notes, just because foundation models work well for coding, That doesn't mean they're ideal for everything else. Here's hoping we take the time as a society to examine our assumptions about this new technology, to take it beyond the realm of engineers and entrepreneurs, but also to use engineers and entrepreneurs in that exploration.

31:41 These are wide open times. and it's only healthy to rely on each other to make it through. I'm Bob Safian. Thanks for listening.

31:58 Rapid response is a wait what original. I'm Bob Safian. Our executive producer is Eve Tro. Our producer is Alex Morris. Associate producer is Mashimaku Tonina. Mixing and mastering by Aaron Bastanelli. Our theme music is by Ryan Holiday.

32:17 Our head of podcast is Lytal Malad. For more, visit rapidresponsshow.com.