AI-Generated Psychotherapy Notes and Assessment
What clients should ask therapists before agreeing to audio recordings and automated records.
Posted January 5, 2026 | Reviewed by Michelle Quirk
Artificial intelligence (AI) is increasingly being used to improve work efficiency, saving time and money. Use of AI by mental health professionals (MHPs) is no exception. One common approach involves audio-recording therapy sessions so the clinician does not have to take detailed notes during the meeting. AI software can then analyze the recording and generate written documentation. Before agreeing to this practice, it is important that you understand how audio-recordings and AI-generated notes may be used, how your information will be protected, and what choices you may have. You may use the following questions to initiate your discussions with your MHP.
What is the purpose of audio-recording?
Audio-recordings can be used for various purposes, so it is important to ask your MHP how exactly they want to use them. One primary purpose may be to have AI create specific types of documentation for the MHP. MHPs are required to maintain accurate documentation of each meeting, sometimes called progress notes. AI tools can convert session recordings into these notes, potentially saving time and allowing MHPs to focus more fully during their therapeutic conversation with you. AI can also create different types of assessments—for instance, screening for suicide or child abuse risks, diagnosing mental health conditions, evaluating service needs, identifying client strengths, and conceptualizing the client’s situation according to particular theoretical frameworks (for instance, cognitive behavioral theory, family systems theory, or social constructionism).
You might agree to some purposes, but not others. Consent does not have to be “all or nothing.” In any case, MHPs have an ethical duty to be transparent about how they are using technology as part of the informed consent process ( ACA, 2014 ; APA, 2017 ; NASW, 2021 ). Your MHP may ask you to sign a written consent form to authorize the use of audio recordings or AI. Review the form carefully. Make sure you understand its terms. Ask questions to clarify anything that you are not certain about. If you are uncomfortable with audio recordings or AI-generated notes, ask about alternatives. For instance, the MHP could simply take notes manually during or after the session.
Who will have access to the audio-recording and AI documentation?
Licensed MHPs have legal and ethical duties to maintain the privacy of all client records, including any audio-recordings and outputs from AI or other digital tools (Barsky, 2023; Reamer, 2025). Under the privacy rule of the federal Health Insurance Portability and Accountability Act (HIPAA), health professionals must protect your private physical and mental health information and not share information with others without your explicit consent (subject to specific exceptions such as the mandate to report child abuse). Find out whether the technologies being used to audio-record and create documentation are “HIPAA-compliant.” HIPAA-compliant means that the technologies meet certain standards for confidentiality—for instance, the use of strong passwords and encryption to prevent unauthorized access to use, a Business Associates Agreement between the MHP and tech provider affirming compliance with HIPAA, and other assurances that information from the audiorecordings and AI will not be shared with marketers, AI developers, or other third parties without the explicit written permission of the client. Your MHP should also be able to explain when and how any data stored on AI systems or other technology will be deleted.
What safeguards will the MHP use to prevent misinterpretation, omissions, or errors in AI-generated documentation?
AI-generated notes and assessments can be subject to several types of errors. For example, AI may misinterpret spoken language when converting audio-recordings into text, particularly when speech is nuanced, emotional, or culturally specific. Some AI systems rely on algorithms that may reflect embedded biases, potentially affecting clinical impressions, assessments, or risk evaluations. AI may also omit important information or, in some instances, generate inaccurate or fabricated details that were never discussed.
If AI is relying solely on an audio-recording, it will not have the benefit of “seeing” your body language , facial expressions, and the social context of your interactions with your MHP. As a result, AI may miss key elements that your MHP would otherwise recognize as clinically significant.
For these reasons, it is essential that MHPs remain fully responsible for the accuracy, completeness, and appropriateness of all clinical documentation, even when AI is used as a support tool. AI is intended to assist professionals, not replace their clinical judgment. Ethical practice requires MHPs to carefully review, edit, and personalize all AI-generated content, correcting any inaccuracies, addressing potential bias , ensuring appropriate language, and tailoring the documentation to your specific situation (The Nerdie Therapist, 2025). In some jurisdictions, laws explicitly prohibit MHPs from relying solely on AI to develop clinical diagnoses or assessments, reinforcing the principle that final clinical responsibility rests with the MHP.
AI has the potential to enhance the efficiency and effectiveness of clinical documentation, assessment, and service delivery. At the same time, its use carries specific risks, often depending on how your MHP implements these tools and, importantly, how they safeguard against concerns such as inaccuracy, bias, missing information, and unauthorized access to your clinical records. From the outset of services, you have the right to know whether and how AI or other technologies may be used in documentation, assessment, intervention, or other aspects of the therapeutic process. Ask questions. Seek clarity. Stay informed. Doing so empowers you to make thoughtful decisions about whether and how AI is being used in your care, ensuring that the technology serves your interests and supports the primary purpose of therapy: your well-being.
American Counseling Association [ACA]. (2014). Code of ethics. https://www.counseling.org/docs/default-source/default-document-library…
American Psychological Association [APA]. (2017). Ethical principles of psychologists and code of conduct. https://www.apa.org/ethics/code
Barsky, A. E. (2023). Essential Ethics for Social Work Practice . Oxford University Press.
National Association of Social Workers. (2021). Code of ethics. https://www.socialworkers.org/About/Ethics/Code-of-Ethics/Code-of-Ethic…
Reamer, F. G. (2025). Artificial Intelligence in the Behavioral Health Professions . NASW Press.
The Nerdie Therapist (2025). AI meets therapy notes: Write smarter, stay ethical, and ditch the burnout. https://thenerdietherapist.com/2025/04/05/ai-meets-therapy-notes-write-…
U.S. Department of Health and Human Services. (2003). Standards for privacy of individually identifiable health information . Federal Register (HIPAA), 68 (34), 8334–8381.
Share this post Facebook Bluesky Linkedin Email
There was a problem adding your email address. Please try again.
By submitting your information you agree to the Psychology Today Terms & Conditions and Privacy Policy
Allan E. Barsky is a professor with the Phyllis and Harvey Sandler School of Social Work at Florida Atlantic University (FAU) in Boca Raton, FL, USA. The views expressed in Dr. Barsky’s blogs do not necessarily reflect the views of any organization to which he belongs.
Get the help you need from a therapist near you–a FREE service from Psychology Today.
This article is part of the Bringwise Psychology Journal — daily insights on human behavior, mental health, and personal growth.