Share this content
Introduction
Artificial Intelligence (AI) is a rapidly developing and transforming technology that has the capacity to impact all areas of our lives. AI has the potential to increase access to mental health services and support, essential in the context of the mental health crisis and lack of providers. For example, fit-for-purpose AI chatbots can reduce self-reported symptoms of anxiety, depression, stress, and loneliness (Chin et al.,2023;, Heinz et al., 2025; Lim et al., 2022; Zhong et al., 2024). AI may help clinicians and researchers examine large samples of individual client data to provide a more tailored, accurate, and timely approach to health care and mental health support, (i.e. Lee et al., 2021; Schueller et al., 2023). AI-powered educational learning platforms are also being developed for students' medical training to improve patient outcomes and reduce health disparities (Bayram et al., 2024; Boscardin et al., 2024). There is well-earned scepticism, too. Advancements in the technology are occurring rapidly and scientific research and regulatory frameworks are struggling to keep up. Emerging literature is highlighting the harms associated with use by certain vulnerable populations (Patel and Hussein, 2024), including children and those with a history of mental health issues (Head, 2025). Furthermore, there is concern that the technology can exacerbate existing inequities and perpetuate bias (Bouguettaya et al., 2025, Timmons et al., 2022; Obermeyer et al., 2019).
Calls to center fairness, ethics, equity, and safety in the design of such systems have been increasing, as have recommendations and frameworks (Gichoya et al., 2021); executing those recommendations is easier said than done. Interdisciplinary differences in definitions of core constructs, research methodologies, training, pace of research, and professional codes of ethics are just a few examples of barriers that impede progress in designing AI for mental health that is fair, equitable and safe.
The purpose of this special issue is to advance our understanding of these barriers and how to overcome them. We welcome submissions that expand and deepen the conversation on equity, safety, and fairness in AI for mental health through empirical testing and evaluation of these concepts. This includes empirical studies that incorporate participatory approaches to system design, as well as detailed descriptions or evaluations of mental health interventions specifically designed with these principles in mind. Conceptual and theoretical papers are also encouraged if they engage comparisons of how constructs such as safety, fairness, and equity are defined and operationalized across disciplines, or non-empirical papers that introduce foundational perspectives from one discipline to foster interdisciplinary understanding and collaboration. For example, the definition and evaluation process of what is considered “trustworthy AI” may vary across disciplines. We also invite contributions that explore the educational and training needs of mental health professionals in this emerging field with regard to safety and fairness principals, and tutorial or methodological papers offering practical tools (e.g., scorable rubrics, tutorials on bias assessment, reproduceable code) to guide future research and practice.
List of Topic Areas
- Evaluation of AI systems in mental health, including studies assessing safety, bias, transparency, and trustworthiness in real-world contexts.
- Therapeutic uses of AI, such as chatbots, conversational agents, or personalized psychological care systems, with particular attention to how safety and fairness are ensured and evaluated.
- AI as a partner in care, (e.g., co-therapy models, human-in-the-loop systems) and empirical assessments of their reliability, safety, and human impact.
- Machine learning and language modelling for prediction and prevention, focusing on how such systems can be tested for bias, equity, and clinical trustworthiness.
- Ethical, practical, and regulatory considerations in developing, implementing, and continuously monitoring AI innovations for mental health care, training, and research.
- Participatory and human-centered methods for designing, auditing, and evaluating AI systems to enhance fairness, explainability, and user trust.
- Development and validation of AI-based diagnostic or assessment tools, emphasizing transparent evaluation of accuracy, safety, and potential bias across populations.
- Educational and professional training uses of AI that prepare mental health professionals to identify, assess, and mitigate risks related to safety and bias.
- Conceptual, theoretical, and methodological innovations, such as frameworks, scoring rubrics, reproducible tutorials, and/or open-source tools for systematically evaluating fairness, safety, and trust in mental health AI.
Submission Information
Submissions are made using ScholarOne Manuscripts. Registration and access are available here:
Author guidelines must be strictly followed. Please see:
Authors should select (from the drop-down menu) the special issue title at the appropriate step in the submission process, i.e. in response to “Please select the issue you are submitting to”.
Submitted articles must not have been previously published, nor should they be under consideration for publication anywhere else, while under review for this journal.
Key Dates
Closing date for manuscript submissions: 31 May 2026
Share this content