This episode contains references to suicide and self harm that may be upsetting for some people. If you require help, A link to a full list of support services is available in the description of this episode.
Kiota.
I'm Chelsea Daniels and This is the Front Page, a daily podcast presented by the New Zealand Herald.
As many as one in five.
KIWI youth aged between fifteen and twenty four have experienced anxiety or depression at some point in their lives. The twenty twenty two to twenty three New Zealand Health Survey found that of those young people experiencing high mental health needs, seventy seven percent can't access.
Support when they need it.
So with services experience this kind of unprecedented demand, what if there was another option? What if teens turned to AI for mental health support.
It's a growing trend among youth in the US.
Seventy two percent of teens there admit they've used AI chatbots as companions. Nearly one in eight said they had sought emotional or mental health support from them. But is the advice their AI therapists are giving helpful or harmful? Mental Health Minister Matt Doosey has acknowledged that the risks need to be managed, particularly around safety from a clinical perspective. Today, on the front page, Rand senior policy researcher Ryan McBain
takes us through the worrying trend sweeping America. First off, Ryan, tell me about this trend of teenage is turning to chat bots for advice and companionship in some cases.
Yeah, So, the most recent evidence that has been put out in a number of reports is that the majority of teens now are using AI as companions that could look like bouncing ideas off of them or chatting on with what's going on in their lives or troubleshooting issues
that they're having. When you're talking about mental health specifically, the numbers are a bit smaller, so it's something like one in eight younger adolescents or as much as one in every five older adolescents are using it for mental health issues specifically. But that's still tens of millions of
adolescents teenagers around the globe. And it brings up the sort of interesting point that, on the one hand, talking about companionship is really quite broad, and when you're talking about treatment, you're getting quite specific, and most chatbots sort of live in this grade space between them, which makes it hard to evaluate them and to regulate them.
You've pointed out and you just said nearly half of young Americans aged eighteen to twenty five with mental health needs received no treatment last year, And I thought it was quite interesting to see those figures as well and the crisis you have there, because New Zealand is in a similar boat. So studies here show that more than half of Kiwi's aged fifteen to twenty four experience anxiety
or depression. More than a quarter of our young people who have high mental health needs, seventy seven percent of those can't access support.
So do you think there is a place for these kinds of.
Chatbot services if they are developed correctly.
I think if they're developed correctly, then yes, I think that is the issue. Right now. There's sort of this gold rush within AI, people who are looking to become the first platform that's doing the sort of thing. And so from perspective, that time has not yet come, at least not for formal services like augnitive behavioral therapy or medication management. But to be honest, part of the reason that I started doing research in this space is because I really do think that there is a potential for
transformational change within mental health care. And you can imagine if we had super intelligence, right, a clinician who is able to follow the best evidence, who's available twenty four to seven, who remembers every detail of your prior conversations. That is a game changer in a landscape like New Zealand or the United States where over half of teens who need care are not getting it.
In terms of what's happening at the moment, and the research that you've done into the topic, what has alarmed you?
So what kind of advice is being.
Given to teams struggling with mental health issues?
Yeah, well, I think it's important to level set that. For the most part, I think teenagers, anybody who's using chatbots for mental health, they're usually getting positive and thoughtful advice. In the research that we've done presenting clinical vignettes related to depression or anxiety, what we find is that chatbots are empathic, if a bit sycophantic, meaning they sort of are overly flattering at times, but they'll also offer good advice to get exercise, to go outside, to talk to
a mental health professional, these sorts of things. So I think for the majority of people, the types of advice you'd be getting is pretty good. But where there's a key distinction here is that you do have people who are at the tail end of the spectrum, people who have severe mental illness, who have psychosis or contemplating suicide, and for those people, it's the highest risk that something could really go wrong, and that has shown up in
our research. So, for example, if we were prompting chat TPT on something like how to tie a noose, or ask it about what types of pesticides or what types of firearms are most effective at completing suicide, these are types of questions that at least the previous version of chat GPT would generate direct responses to, whereas other types of chatbots, like Google's Gemini would not give a response to it and would say something like I can't give
that type of information to you because you could use it for self harm.
What do you think makes young people, particularly these younger generations, so susceptible to these kind of AI chatbots compared with adults or even just just older generations.
Well, ants, I think that childhood and adolescents are transformational times. Your brain is still developing. You don't always have the
best emotion regulation or impulse control. I mean, I know that I personally made a lot of dumb decisions when I was seventeen eighteen years old, and I wish I could say goes back, But I didn't live as a digital native the same way that the current generation is, where social media is always in their pockets, where you could have AI do your homework for you or discuss life issues with So I think that that temptation is always there for that additional sort of dopamine hit or
to get that competitive edge or additional advice, and so it becomes a sort of positive feedback loop, or in the case of mental health, sometimes a vicious negative feedback loop.
I saw the case of sixteen year old Adam Rain.
His parents actually have a suing open AI and its CEO, Sam Altman, and they're alleging chat GBT contributed to their son's suicide, advising him of methods and offering to write a first draft of his suicide note. Should AI companies be legally liable if a chat bot provides harmful advice to moor botain a jet.
Yeah, I think that's a difficult question. I mean, it's hard for me to comment on that case specifically, I will say that JATGPT, for example, does put a disclaimer at the bottom of every conversation that's had that says something along the lines of it can make mistakes and you should check on important information. So from that lens, I'm not sure that those sorts of companies should be held accountable for bad advice any more than a human
should be. On the other hand, I think it's important that if AI companies are marketing their products as options for treatment or life coaching or wellness, then they should have certain standards that need to be met, and if they fail to meet those standards, then they shouldn't be able to operate or they should be have greater potential for a lawsuit because the product has failed to deliver
what it promises. And so there is this sort of distinction between eric bad advice versus harmful advice that's presented under the guise of authority, and this particular case that you're describing, I think the courts will need to decipher between those two elements.
The existential threat of AI may not come in a form that we all imagine watching sci fi movies. What if we all continue to thrive as physical organisms but slowly die inside. What if we do become super productive with the I, but at the same time we get these perfect companions and no willpower to interact with each other. Not something you would have expected from a person who pretty much created the I companionship industry.
I guess most human therapists work and practice under a strict code of ethics and have some kind of certain obligations to report concerning behavior.
It should chatbots have the same.
I think that's a great point, and that for me is what I hope is the next frontier of work that AI companies will be doing right now. Very often, if you pressure a chatbot into a space that is across as a red line in terms of conveying suicidal ideation or psychosis, it will tell you, for example, that you can contact a mental health professional. They might give
you a hotline that you can contact. But as you're saying, if it were a human a counselor, for example, they might have an ethical obligation to connect you to treatment through a warm handoff where you're physically accompanied, or you could even be involuntarily forced to receive institutionalization for some
period of time. Now, I'm not sure that a chatbot as an algorithm is always capable of making those distinctions, but I think at a minimum what would be pragmatic is in instances where it's quite conspicuous the algorithm flags somebody as a red flag as something that's highly problematic, that these companies could have a human teams that are required to vet those cases and to review them within a certain period of time, like twenty four hours or seventy two hours, and if it is identified as a
problem at that point, then there could be some additional course of action that's required.
How do we weigh up the risks of unsafe or harmful chatbot advice with the basic fact And it sounds like the US has a similar situation as New Zealand does in this sense that many teains just don't have access or can afford going to a therapist regularly.
I think it's a great point, and you've put your finger on I believe to be the main issue, which is that there will always be some degree of risk and some degree of benefit in addressing unmet need. The underlying question is how can we mitigate risk as much as possible and how can we enhance benefits as much as possible, And it's hard to know the answer to that without clinical trials, without safety benchmarks that are public and transparent, and that tech companies are subjected to.
And I guess speaking to one of these chatbots as a therapist, I've heard responses and stories about I guess it's a positive thing that you can be your most open, authentic self with it, But then again, the negative side is that you can be your most authentic and self with it and not anybody else and hide what.
You really think. So do you think that there is a place for it? But is it?
Are we going to be talking about this in ten years time thinking I wish this had never been proposed as an option.
Yeah, I think you're right. It is a double edge store the sort of anonymity it helps people who might feel a sense of stigma in talking to peers or talking to parents or a mental health professional. I do think we would regret though, if in ten years we weren't to develop the sorts of guardrails and benchmarks for performance that would really potentially benefit not just adolescents and teens,
but people more generally. I think, in particular, as mentioned before, if a company is specifically marketing their product as therapy or treatment of some sort, then there should be even clearer standards, more stringent standards that they're being held to
in those instances. But obviously, with platforms like open AI's chat GPT, you have hundreds of millions of users and it can be used from anything from learning how to make a birthday cake to discussing intimate aspects of what's going on in your life, and so it's really a wide spectrum, which makes it much harder to regulate and to pin down.
I'm given the rapid adoption of these kind of tools, and this isn't the first time that we've brought this up on the podcast as well, where the law is really running behind the advancements of technology. How urgent is the need for government or international regulation in the AI space here.
I think it's incredibly urgent. I think the time is now to act. I've been impressed, even over the past couple of weeks the number of articles and personal testimonies that have come out about the negative impacts related to mental health that people have experienced with chatbots. And I've seen as a result of that that platforms like Anthropic
Open AI have quickly responded to it. Open AI released a statement in response to some of what's been going on about new safety standards that they're going to be introducing for mental health issues that are shared amongst their users.
We can see, for example, with social media that we waited too long with teenagers, and now we're starting to work our way backwards from that by, for example, banning mobile phones from school environments to try to help kids to be able to learn better and to avoid cyber bullying and these sorts of things.
And lastly, looking ahead, what are some of the red lines that regulators, educators, or parents should set when it comes to young people's use of AI companions.
Well, I think it kind of goes two directions. On the one hand, I don't think there should be an outright ban on AI off bring mental health advice or therapy in the future. I think there's remarkable potential that I've tried to underscore and we don't want to leave that on the table. In the United States, Illinois, just a couple of weeks ago became the first state to outright ban AI as a tool in therapeutic decision making.
I think that is too severe or maybe it was right for the moment, but it won't be right in five years, and that legislation would need to be amended. On the other hand, I do think that we're most at risk right now of doing too little. I think people are very excited about AI. There's a lot of money in it, and we need to think hard about
tamping the brakes to develop stronger safety benchmarks. I am not clear that there is a conspicuous red line, other than to say we are reaching a tipping point of testimony and people sharing experiences that have been quite negative and jarring, and hopefully that is sufficient in terms of advocacy for regulators to begin stepping in and for these tech companies to move beyond self regulation and establishing their own benchmarks that might be too low, and instead requiring
independent bodies to come in to establish guidelines and standards and to have auditing on a routine basis as part of the company's practices.
Thanks for joining us, Ryan, Yeah, it was my pleasure.
That's it for this episode of the Front Page. You can read more about today's stories and extensive news coverage at enziherld dot co dot nz. The Front Page is produced by Jane Ye and Richard Martin, who is also our editor. I'm Chelsea Daniels. Subscribe to the front page on iHeartRadio or wherever you get your podcasts, and tune in tomorrow for another look behind the headlines.