Professional Ethics, Artificial Intelligence (AI), and Clinical Practice

Few topics have generated as much discussion, concern, media attention, and buzz in recent years as the development and implementation of Artificial Intelligence (AI) in our personal and professional lives and in our global society as well. Although broadly available for just a few years, we have already been greatly impacted by AI and will likely continue to be influenced in numerous and sometimes unpredictable ways moving forward. AI has upended so many aspects of life, presenting both opportunities and risks.

Warnings and rumors about AI range from the use of autonomous weapons in war, to the generation of realistic sounding misinformation, to academic cheating, job eliminations, and environmental damage associated with powering large data centers across the country and world. Remarkably, Pope Leo XIV focused his first encyclical letter, Magnifica Humanitas, On Safeguarding the Human Person in the Time of Artificial Intelligence on how AI might impact our global society and especially those who are the most vulnerable and marginalized in our communities (Pope Leo XIV, 2026). Like any new technology or innovation, AI can be used for good purposes or for bad ones and so considering ethical implications in the use of AI would be helpful for everyone, in every industry, including our own (Bellaby, 2024; Floridi, 2024).

Board specialists and clients are likely experiencing some struggles and confusion regarding the use of AI in our professional work. For example, AI can be used to generate psychological testing reports and evaluations from assessment data that we collect. In fact, AI tools have been used in test interpretation software for many years, and many psychologists may not be aware of this. Both clients and professionals may then put these reports into AI searches to better understand the assessment findings. Many of our psychotherapy clients turn to AI for psychological help and even ongoing therapy with AI chatbots and other newly developed AI products and services (Bhatt, 2025; Tekin & Delehanty, 2026). Board specialists who write books, journal articles, or conduct empirical research may use AI to improve their writing, generate their prose, and better understand their results from their data collection. Specialists who teach may find that their students and trainees use AI to generate their papers, reports, or examination answers. Since AI products, services, and uses are developing and advancing rapidly, it is difficult to keep up with the advances and develop clarity about how AI should or should not be used in our professional work.

Our APA Code of Ethics (American Psychological Association, 2017), although published long before AI entered our imagination and the global scene, offers some helpful wisdom about how we should approach AI in our professional work. There are several key principles that we should keep in mind as we try and navigate the shifting sands and new developments in this remarkable game-changing technology. Perhaps the most important ethical principles and guidelines from our Code of Ethics to consider and highlight in this brief article include (1) Section 4 (Privacy and Confidentiality), (2) Section 3 (Human Relations), and (3) Section 2 (Competence). Let us review each of these three sections and speak to how they may relate to the use of AI in our professional work.

Section 4: Privacy and Confidentiality

4.01 Maintaining Confidentiality
Psychologists have a primary obligation and take reasonable precautions to protect confidential information obtained through or stored in any medium, recognizing that the extent and limits of confidentiality may be regulated by law or established by institutional rules or professional or scientific relationship.

4.02 Discussing the Limits of Confidentiality
(a) Psychologists discuss with persons (including, to the extent feasible, persons who are legally incapable of giving informed consent and their legal representatives) and organizations with whom they establish a scientific or professional relationship (1) the relevant limits of confidentiality and (2) the foreseeable uses of the information generated through their psychological activities. 

Care is needed when AI is used with clinical material, so that confidentiality is not compromised, unless a HIPAA compliant AI tool is used, or the AI tool is run locally so that data privacy is preserved. If AI is used with clinical material in a way where that clinical material is not safeguarded, AI companies may use this information to further develop their large language models, or for other purposes. Similarly, professionals and clients need to fully understand that confidentiality is likely compromised when they use therapy chatbots or when clinical material is uploaded into many AI platforms. Adding to the confusion, product disclosure practices vary widely. Conducting due diligence often requires getting guidance on assessing competing technical and contractual AI safeguards (e.g., data-retention policies), since these skills are outside our current toolbox.

Section 3: Human Relations

3.04 Avoiding Harm
(a) Psychologists take reasonable steps to avoid harming their clients/patients, students, supervisees, research participants, organizational clients, and others with whom they work, and to minimize harm where it is foreseeable and unavoidable. 

It is critically important that we avoid harm. Perhaps this is our first and most important ethical duty in our professional work. Yet, harm can be caused by confidentiality breaches or when using psychotherapy chatbots and services to seek help for psychological and relational distress. For example, there have been reports of people using AI therapy platforms and services, especially young people, who later die by suicide or develop worsening eating disorders (Bunch & Noble, 2026; Tekin & Delehanty, 2026). Most of these AI therapy services are not developed by licensed mental health professionals but by engineers and business professionals who have limited if any mental health training, experience, or licensure. They wish to develop these products and services quickly to get them to market without standard research processes like conducting randomized clinical trials to demonstrate effectiveness and safety. These companies are also not beholden to ethical principles, such as ours.

3.10 Informed Consent
(a) When psychologists conduct research or provide assessment, therapy, counseling, or consulting services in person or via electronic transmission or other forms of communication, they obtain the informed consent of the individual or individuals using language that is reasonably understandable to that person or persons except when conducting such activities without consent is mandated by law or governmental regulation or as otherwise provided in this Ethics Code.

Many ethical challenges and conflicts can be resolved by full and informed consent. Laying out and seeking consent in the case of clinical data input into AI by professionals and clients is necessary. Most journal and book publishers now demand author disclosures about AI use in the generation of articles and books with specific guidelines about their use and potential misuse. For example, it is increasingly common that psychological reports and evaluations as well as progress notes may need to disclose if AI has been used and in what ways.

Section 2: Competence

2.01 Boundaries of Competence
(a) Psychologists provide services, teach, and conduct research with populations and in areas only within the boundaries of their competence, based on their education, training, supervised experience, consultation, study, or professional experience.

Most board specialists are not experts in technology or AI products and services. They must be careful about engaging in using and promoting AI in their professional work if they do not have the competence to do so. Obtaining appropriate consultations from colleagues who do have expertise in this area is suggested. Fortunately, our professional organizations, such as the APA, often offer training webinars on a regular basis to address these issues.

It is also important that psychologists do not utilize or rely on AI to extend their scope of practice into areas that they do not have sufficient competence to practice in. It may be tempting, for instance, to think that with the assistance of AI, one could enter test data and generate a neuropsychological report, without formal neuropsychological training. We must continue to practice within the bounds of our competence and not allow the ease and availability of AI to falsely lure us into thinking one’s competence or scope of practice has expanded.

Conclusion

Although new and developing, AI is here to stay and will continue to be part of our professional and personal lives in the foreseeable future. Like most technological advances and inventions, there are both positive and negative implications. There may be unintended problematic consequences likely to unfold in due time. Board specialists may wish to proceed in using AI with caution and with our ethical principles always in mind. Our ABPP Ethics Committee is always available for consultation and we welcome questions and engagement with all our board specialists. See here for details: https://abpp.org/specialists/ethics-education-consultation/ethics-consultation-process/. 

While we may not have all the answers and we cannot offer legal guidance, we can hopefully assist in thinking through the ethical issues and implications in a timely, thoughtful, and collegially manner.