CBT Trainer: An AI Virtual Patient Platform for Training CBT Practitioners against Competencies — Pilot Study

Background: CBT training faces significant challenges including limited supervised practice opportunities with diverse cases, inconsistent training experiences, resource-intensive supervision methods, and difficulties ensuring standardised competency assessment. This study evaluated the acceptability and feasibility of CBT trainer, an app facilitating training of CBT therapists through AI simulation of therapeutic interactions with feedback supporting competency development.Method: This mixed-methods pilot study employed a two-stage approach. Stage 1 involved initial usability testing. Stage 2 included 59 participants from psychological practitioner training programs (77.2% from Low Intensity CBT Interventions program, 22.8% from Doctorate in Clinical Psychology) who engaged with the application over a month. Outcome measures included the System Usability Scale, platform engagement, post-study surveys on perceived learning outcomes, and qualitative feedback.Results: The CBT Trainer exceeded all pre-specified outcome targets, with participants spending an average of 95.24 minutes (SD = 134.58) engaging with roleplays. Participants voluntarily created an average of 4.15 role-play sessions each (SD = 3.55), with each session containing an average of 49.69 interactions. The platform achieved excellent usability with a mean SUS score of 82.20 (SD = 12.93). Assessment skills were the most significantly self-reported improved area (96.7%), followed by information gathering skills (66.7%). Participants valued immediate feedback (83.3%) and convenience (73.3%) as key advantages over traditional methods. Conclusion: This pilot study provides evidence that AI-based virtual patient simulation offers a promising supplementary training tool for CBT therapists, with particular strengths in accessibility and immediate feedback. Future research should employ randomized controlled designs with objective competency assessments to determine whether simulated practice translates to measurable improvements in clinical performance.