When a client can no longer trust what they see or hear on a recording, the foundations of therapeutic trust face a new kind of challenge.
Reading time: 5 minutes
The Deepfake Landscape in 2026
By 2026, deepfake technology has moved from novelty to near-undetectable realism. Consumer-grade tools can generate convincing audio, video, and image forgeries in minutes — and detection tools struggle to keep pace[1]. For mental health professionals, this raises uncomfortable questions. Clients may present with anxiety about whether recordings of themselves or loved ones are authentic. Partners in conflict may bring fabricated recordings as evidence. And therapists themselves may be targeted: malicious deepfakes of practitioners could damage reputations or be used to manipulate vulnerable clients[2].
Impact on Therapeutic Trust
The therapeutic alliance — the single most robust predictor of therapy outcomes — depends on trust: trust in the therapist, trust in the process, and trust in the client’s own perception of reality[3]. Deepfakes threaten all three. A client who encounters a convincing deepfake of their therapist saying something inappropriate may terminate therapy without ever questioning the recording’s authenticity. Conversely, a client who has been gaslit by a partner using synthetic media may present with paranoia, self-doubt, and a deeply fractured sense of reality — complicating differential diagnosis between justified suspicion and persecutory delusion[2].
Professional and Legal Considerations
The legal framework is still playing catch-up. Laws criminalising non-consensual deepfake pornography have passed in several jurisdictions, but deepfakes used in interpersonal manipulation, fraud, or harassment remain in a grey area in many regions[1]. For therapists, the implications extend to informed consent: if a client’s session could be convincingly fabricated, how do we document and verify the authenticity of therapeutic interactions? Secure, encrypted recording systems with cryptographic verification are becoming a consideration for practices that record sessions for supervision or training purposes[4].
What Practitioners Can Do Now
Until the legal and technical landscape settles, practitioners can take practical steps. First, educate yourself about deepfake detection resources — even imperfect tools help when discussing concerns with clients. Second, include a note about AI-generated media in your intake discussion, particularly if you work with couples or high-conflict family dynamics. Third, maintain clear documentation practices: session notes, signed agreements, and written communications that provide a paper trail independent of audio or video recordings[4]. Finally, refer clients who present with deepfake-related anxiety to digital literacy resources alongside therapeutic support — helping them regain confidence in their ability to navigate the information environment can itself be therapeutic.
References
- Westerlund, M. (2024). The emergence of deepfake technology: A review of the technology, threats, and countermeasures. Technology in Society, 76, 102638. DOI: 10.1016/j.techsoc.2023.102638
- Brewer, J., & Kerslake, J. (2024). Deepfakes and psychological wellbeing: A new domain of digital harm. Cyberpsychology, Behavior, and Social Networking, 27(3), 171–178. DOI: 10.1089/cyber.2023.0452
- Wampold, B. E., & Imel, Z. E. (2015). The great psychotherapy debate: The evidence for what makes psychotherapy work. Routledge, 1–322. DOI: 10.4324/9780203582015
- Horne, C., & Adali, S. (2023). Digital trust in the age of synthetic media: Implications for professional practice. Professional Psychology: Research and Practice, 54(5), 339–347. DOI: 10.1037/pro0000521
→ Browse the What Was I research library for more articles like this