The U.S. is grappling with a significant shortage of mental health professionals. According to a recent survey by HRSA, over 60 percent of therapists cannot accept new patients due to high demand. As a result, one in three individuals today are left waiting for months before accessing care, with low-income areas and communities of color being the most affected. Moreover, challenges including high costs of therapy, limited availability and social stigma toward mental health care compound these issues.
Aiming to fill some of the shortcomings in the space, A.I. mental health care companies are cropping up, claiming to make mental health care better and more accessible. The New York-based mental health chatbot Slingshot AI recently raised $30 million from Andreessen Horowitz, and A.I.-powered Spring Health raised a $100 million Series E in July, signaling Silicon Valley’s excitement for the technology’s potential in the mental health space. The A.I. mental health market is expected to reach $12.67 billion by 2031, according to a recent report by Netscribes.
One of the big questions surrounding the use of A.I. in mental health care is whether it can truly replicate the empathy and trust that human therapists provide. While the technology can offer consistency in care and eliminate biases affecting human therapists, it still lacks the genuine emotional connection that many patients find important, especially in therapy.
However, A.I.-generated communications “demonstrated superior discipline in offering emotional support” to recipients compared to untrained human interactions, according to a new study published in the Proceedings of the National Academy of Sciences (PNAS).
The study also suggests that A.I. has the potential to offer a sense of acknowledgment and understanding through sophisticated algorithms and natural language processing. In another study by researchers at Columbia University published last month, nearly half of patients surveyed said they “believed A.I. may be beneficial for mental health care.” However, participants noted concerns over potential for misdiagnosis, patient data security, as well as “loss of connection with their health professional.”
Citing the current state of health care in the U.S., some experts argue that having A.I. attending patients’ mental health care needs is better than having no one at all. “In the Bay Area, it takes months just to schedule a first therapist appointment. Visibility to such bottlenecks is essential to enhance triage processes and get individuals to the correct level of care,” Grace Chang, the founder and CEO of Kintsugi, told Observer. Her company develops voice biomarker A.I. technology to detect signs of depression and anxiety from short clips of speech in real time.
She said technology is a powerful tool for providing care to underserved populations. “Since A.I. can be deployed on a large scale at a lower cost, it has the potential to reach millions of people simultaneously, addressing the demand and supply issues in mental health care.”
A.I.’s role in mental health care goes far beyond improving therapeutic accessibility. Social stigma is often a major to mental health care, discouraging individuals from seeking help. By analyzing speech patterns, genetic information and lifestyle choices, A.I. models can tailor treatments to the specific needs of individuals, enabling privacy in the comfort of their homes.
For example, A.I. models can predict suicide attempts up to a week in advance with 92 percent accuracy, due to the technology’s ability to detect nuanced patterns in speech and behavior, according to a study by NCIB. These approaches could in-turn enhance therapists’ effectiveness while minimizing the trial-and-error processes often associated with mental health treatment.
“When mental health issues are discussed in a medical environment with a few additional, non-stigmatizing questions, it opens up opportunities for individuals to receive the appropriate level of care,” Chang said. But of course, not all patients fully communicate their feelings all the time. Grace added that deep learning models that can analyze speech patterns for signs of depression within seconds can help nurses and practitioners treat patients during telehealth sessions, even in the face of stigma.
Ultimately, the success of A.I. in mental health care may hinge on its ability to integrate into existing health care systems, maintain ethical standards, and continually improve its empathy and understanding capabilities. While it may not replace the empathetic understanding of human therapists, the technology has the potential to play a pivotal role in enabling more individuals to receive the required care.
