‘How do you use AI?’ Experts say therapists should ask you: NPR

Recent studies and surveys show that teens and adults are increasingly turning to artificial intelligence chatbots for peer interaction and emotional support. Therefore, mental health care providers should investigate whether and how their patients are using this technology, just as they would ask for information about sleep, diet, exercise, and alcohol intake.

According to a new paper, JAMA Psychiatry.

“We are not saying that using AI is good or bad,” Shadi Saba says. “Just as drug use is not necessarily good or bad, [or] To consult a friend about something good or bad. ”

But learning about people who rely on AI for emotional support and advice could provide valuable insight into their lives and mental health, he says.

“Our job is to understand why people are behaving the way they are, and in this case, why they are turning to AI systems for help,” Saba adds. “And to find out what it’s doing for them and what’s not working for them.”

Saba and his co-authors’ recommendations are “very consistent” with those recommended by the American Psychological Association (APA) in its health recommendations released last November, APA’s Vail Wright said.

Asking patients what they’re getting out of conversations with AI chatbots is “the foundation for therapists to better understand their patients’ mental health and how they’re trying to navigate mental illness,” Wright says.

“Treasure trove of information”

“People regularly use these tools to ask about ways to cope with stressful experiences and relationship challenges,” Saba explains.

Others are using chatbots to get advice on how to deal with anxiety and depression symptoms.

“I think there’s a potential treasure trove of information if we can encourage clients to have these conversations in more detail and in the treatment room,” he says.

It could be information about the main sources of stress in someone’s life, or whether they rely on chatbots as a way to avoid conflict.

“For example, let’s say you have a client who is having trouble with their spouse,” says APA’s Wright. “And instead of trying to have a frank conversation with their spouse about how to meet their needs, they are instead using chatbots to meet their needs or avoid having difficult conversations with their spouse.”

She explains that this background helps therapists better support patients.

“We help you understand how to have safe conversations with your spouse and understand the limitations of AI as a tool to bridge the needs gap.”

Discussing the use of AI is also an opportunity to learn about things clients may not voluntarily share with their therapists, says Dr. Tom Insel, a psychiatrist and former director of the National Institute of Mental Health. “People often use chatbots to talk about things they can’t talk about to other people because they’re worried about being judged,” he says.

For example, suicidal thoughts may not be something a patient wants to share with their therapist, but it’s important for the therapist to know to keep the patient safe.

Be curious but don’t judge

Saba recommends first broaching the subject with patients without judgment.

“We don’t want our clients to feel like we’re criticizing them,” he says. “If we did that, they generally wouldn’t want to work with us.”

He recommends that therapists approach this topic with genuine curiosity and provide suggested language for these conversations.

“‘You know, AI is something that’s growing rapidly, and I’m hearing from a lot of people that they’re using things like ChatGPT for emotional support,'” he suggests. “‘Does that apply to you? Have you tried it?'”

We also encourage patients to ask specific questions about what has been helpful to them to better understand how they are using these tools.

Insel says it could also help therapists determine whether chatbots could complement therapy in a beneficial way, such as scrutinizing what topics to bring to a session or venting about daily life.

In some ways, therapy and chatbots “could work hand in hand,” Insel said.

Saba and co-author William Weeks also suggest asking patients if they found chatbot interactions unhelpful or problematic, and sharing the risks of using chatbots for emotional support.

For example, many AI companies use conversations (even sensitive ones) to further train their models, which poses risks to data privacy.

There are also risks to treating chatbots like therapists, Insel says.

Talking to a chatbot about your mental health is “the opposite of therapy,” he says. Chatbots are designed to affirm and flatter users’ thoughts and feelings.

“Therapy is there to change you, to challenge you, and to get you to talk about especially difficult things,” Insel says.

take advice

Psychologist Cami Winkelspecht runs a private practice in Wilmington, Delaware that primarily works with children and adolescents.

She was considering adding questions about the use of social media and AI to her intake form and appreciated Saba’s work for providing some sample questions.

Over the past year or so, Winkelspecht has seen an increase in customers and their parents asking for help on how to use AI for brainstorming and other tasks in a way that doesn’t violate their school’s honor code. So she had to become familiar with technology so she could support her clients. Along the way, she realized that therapists and parents of children need to be more aware of how children and teens use digital devices (both social media and AI chatbots).

“We don’t necessarily think that much about what they’re doing on their phones,” Winkelspecht says. “And I think it’s clear that we need to do more of that and encourage ourselves to have those conversations.”