Summary:
A survey study led by researchers at Tel-Hai University of Kiryat Shmona reveals that almost one in three Israeli adults using general-purpose generative AI tools rely on platforms like ChatGPT and Gemini for emotional support, coping advice, or mental health assistance. Published in Cyberpsychology, Behavior, and Social Networking, the investigation found that among individuals experiencing severe psychological distress, the adoption rate surged to 47%. The findings highlight a growing trend of informal, AI-mediated help-seeking, raising critical questions regarding clinical safety, crisis protocols, and the regulation of consumer AI systems.
Key Facts:
- High Prevalence in Severe Distress: While 29.3% of surveyed generative AI users reported seeking emotional or mental health support from AI platforms at least once in the past year, that rate climbed to 47% among those reporting severe psychological distress.
- Demographic Patterns: Severe psychological distress emerged as the strongest predictor of turning to AI for emotional guidance, with younger adults and women significantly more likely to utilize conversational models for mental health support.
- Absence of Clinical Mandates: General-purpose commercial AI platforms lack formal psychiatric validation, clinical mandates, and standardized safety protocols to manage acute psychological crises or self-harm risks.
Source: Tel-Hai University of Kiryat Shmona
As generative artificial intelligence tools become embedded in daily workflows, millions of users are discovering that large language models offer far more than text generation and software coding: they provide a conversational partner that is always available, endlessly patient, and completely nonjudgmental.
While tech developers build these general-purpose chatbots for informational, educational, and creative tasks, individuals are increasingly co-opting them for psychological care.
Now, an empirical investigation led by Dr. Dorit Hadar Shoval of the School of Psychology at Tel-Hai University of Kiryat Shmona in the Galilee, in collaboration with Elad Refoua from Bar-Ilan University, documents the scale of this digital help-seeking shift.
Publishing in the peer-reviewed journal Cyberpsychology, Behavior, and Social Networking, the researchers examined how psychological distress and interpersonal attachment styles correlate with utilizing consumer chatbots for mental health guidance.
Psychological Distress Drives Chatbot Consultation
The research team surveyed 584 Israeli adults, aged 18 to 70, who reported using a consumer generative AI tool—such as OpenAI’s ChatGPT or Google’s Gemini—at least once during the previous 12 months.
The survey revealed that nearly a third of all respondents (29.3%) turned to generative AI for coping advice, emotional reassurance, or mental health-related assistance. Among individuals suffering from severe psychological distress, however, the proportion escalated to nearly one in two (47%).
Statistical modeling confirmed that the intensity of psychological distress was the single strongest predictor of seeking solace from an algorithm. In addition, younger users and female respondents were significantly more likely to confide in AI platforms regarding personal and psychological struggles.
“Our findings show that many people, particularly those experiencing significant distress, are turning to generative AI systems for emotional support and guidance,” explained Dr. Hadar Shoval. “These tools offer immediate, accessible, and nonjudgmental interactions, but they were not designed to function as mental health providers. The high level of use among distressed individuals highlights the urgent need to better understand both the possible benefits and the potential risks of relying on AI during periods of emotional vulnerability.”
The researchers also analyzed attachment styles, observing that adults with elevated attachment anxiety were initially more inclined to consult AI for emotional reassurance. However, once baseline psychological distress was integrated into the regression models, attachment anxiety ceased to be an independent predictor, suggesting that acute emotional suffering drives the immediate impulse to seek digital companionship.
The Unregulated Frontier of AI-Mediated Help-Seeking
The rapid adoption of conversational agents as informal therapeutic proxies highlights a glaring regulatory and clinical void.
Dedicated digital therapeutics and licensed teletherapy platforms are bound by strict medical device standards, patient confidentiality rules, and emergency escalations for suicidal ideation. In contrast, commercial consumer LLMs carry no clinical mandate. Because these platforms are trained primarily on broad internet text to produce statistically plausible and agreeable responses, their outputs may lack clinical validity, foster unhealthy emotional dependencies, or deliver inappropriate guidance during acute psychological crises.
The authors caution that their study surveyed an online panel of existing AI users rather than a randomized epidemiological cross-section of the general population, and the data do not evaluate whether these AI interactions proved therapeutic, neutral, or harmful over time. Furthermore, the findings do not determine whether users turned to AI as a temporary supplement to human care or as a total replacement for unaffordable or unavailable professional psychiatric services.
Next Steps for Clinical Practice and Policy
With waiting lists for mental health services swelling worldwide, conversational algorithms are quietly filling a massive public health vacuum.
“Understanding patients’ digital help-seeking behaviors is becoming an important part of modern mental health care,” Hadar Shoval stressed. “Future research should focus on safety, crisis-response quality, and whether AI serves as a complement to professional support or, in some cases, replaces it.”
The authors conclude that mental health clinicians must begin routinely inquiring about their patients’ digital habits, while developers and policymakers must establish standardized clinical guardrails, ensuring that emotionally vulnerable individuals confiding in algorithms are directed toward verified, evidence-based human care.
Editorial Notes:
- This article was edited by a Neuroscience News editor.
- Journal paper reviewed in full.
- Additional context added by our staff.
About this AI and mental health Research:
- Media Contact: Einat Paz-Frankel
- Source: Tel-Hai University of Kiryat Shmona in the Galilee
- Image Credit: Image credited to Neuroscience News
- Original Research is Open Access: Cyberpsychology ,Behavior, and Social Networking (Oct 1, 2026). “AI-Mediated Mental Health Support: The Role of Attachment Orientation and Psychological Distress.” Authors: Dorit Hadar Shoval, Elad Refoua, Karny Gigi, Yuval Haber, Inbar Levkovich, Iftach Tsafrir, Ori Pen, and Zohar Elyoseph.
- DOI: 10.1177/21522715261494423
Abstract
AI-Mediated Mental Health Support: The Role of Attachment Orientation and Psychological Distress
General-purpose generative AI (GenAI) models are increasingly serving as quasi-social agents, representing a novel form of informal AI-mediated help-seeking.
This study examined associations between adult attachment orientations (Experiences in Close Relationships-Relationship Structures questionnaire), psychological distress (Depression Anxiety Stress Scales–21), and the use of GenAI for mental health support.
In a nonprobability sample of 584 Israeli adult GenAI users, 29.3 percent reported having turned to these models at least once for emotional support, coping advice, or mental health assistance; the proportion reached 47.0 percent among those reporting severe distress.
These self-reported figures reflect any such contact, from a single exchange to sustained use. Hierarchical logistic regression identified distress as the strongest psychological correlate of use. Younger age and female gender predicted use.
Attachment anxiety was associated with use, but this association did not remain significant once distress was included, a pattern consistent with the shared variance between the two constructs. Attachment avoidance was negatively but nonsignificantly associated with use.
Findings position general-purpose GenAI as an accessible, low-barrier resource disproportionately used by distressed individuals. The high uptake among severely distressed users (47.0 percent) points to potential safety concerns and to a need for systematic evaluation of the safety, crisis-response quality, and clinical-displacement risks of consumer-facing AI use.
