AI-Generated Deepfake Therapy for Trauma Survivors

Marina vd Roest is among the first participants in an experimental therapy that utilizes AI-generated deepfakes of abusers to help survivors of trauma confront their past. This radical method aims to facilitate healing for individuals suffering from post-traumatic stress disorder (PTSD) by allowing them to engage in simulated conversations with lifelike representations of their attackers. The therapy is currently being studied in a larger trial in the Netherlands, with results expected next year.

Slate

During the sessions, survivors are guided by therapists who operate the deepfake technology. The process begins with thorough preparatory interviews to establish the survivor's goals and expectations. Participants are encouraged to bring photographs of their abusers, which are used to create realistic avatars that respond to the therapist's movements. This setup allows survivors to express their feelings and confront their attackers in a controlled environment. For instance, vd Roest described her experience of questioning the deepfake about her abuse, expressing anger and pain that had been suppressed for years.

Slate

The therapy aims to empower survivors by allowing them to hold their abusers accountable. Therapists are trained to adopt different personas based on the nature of the abuse, ensuring that the survivor feels safe while confronting their trauma. The sessions are designed to be limited in number, with some participants finding closure after just one encounter, while others may require multiple sessions to fully engage with the process.

Slate

While the therapy shows promise, there are concerns about the potential for re-traumatization. Marieke Bak, an assistant professor in medical ethics, acknowledges that the risks are theoretically possible but emphasizes that preparatory sessions are crucial in mitigating these dangers. The therapy is conducted in a clinical setting, which is deemed safer than unsupervised attempts at similar interventions. Ethical discussions also revolve around the privacy of perpetrators, as their images are used without consent. Bak argues that the needs of patients with severe PTSD often outweigh the privacy concerns of the abusers.

Slate

Initial findings from the pilot study indicate that participants, including vd Roest, experienced a reduction in PTSD symptoms and increased self-forgiveness shortly after the sessions. However, comprehensive results from the ongoing study will provide a clearer picture of the therapy's effectiveness. As the field of AI therapy evolves, researchers are also considering the implications of deepfake technology in mental health treatment, including the potential for misuse outside of clinical settings.

Slate

• “For the first time in all those years I could express my feelings, my anger, my pain,” — Marina vd Roest, Therapy Participant

Slate

• “It’s really about having good therapists in control,” — Marieke Bak, Assistant Professor in Medical Ethics

Slate

• “If someone has a complicated case of PTSD, and this is thought to be the thing that will work … we say that the legitimate interest of the patients would generally outweigh the privacy risks of the perpetrator,” — Marieke Bak, Assistant Professor in Medical Ethics

Slate

This innovative approach to therapy represents a significant development in the treatment of trauma, blending technology with psychological healing while navigating complex ethical landscapes.

Slate