# Expert-rated quality of AI-assisted clinical supervision in strategic psychotherapy: an ongoing pilot study - URL: https://alisanburak.com/en/wcp-2026-posterler/5619 - Language: en - Last updated: not specified - Page title: Expert-rated quality of AI-assisted clinical supervision in strategic psychotherapy: an ongoing pilot study | WCP 2026 Summary > Blinded expert supervisors found 89% of the supervision feedback generated by a Spanish-language systemic strategic AI assistant clinically appropriate and 91%… This page was translated automatically from Turkish. [Read the original](https://alisanburak.com/wcp-2026-posterler/5619) This page is an AI-assisted Turkish summary of the study presented at the 26th World Congress of Psychiatry. The study belongs to the presenting authors; please refer to the original source when citing it. Brief Oral Poster PresentationDigital Psychiatry and Artificial IntelligencePilot study # Expert-rated quality of AI-assisted clinical supervision in strategic psychotherapy: an ongoing pilot study SO245A - EXPERT-RATED QUALITY OF ARTIFICIAL INTELLIGENCE-ASSISTED CLINICAL SUPERVISION IN STRATEGIC PSYCHOTHERAPY. AN ONGOING PILOT STUDY (ID 5619) Presenter: Sergio A. Grosman Country: Argentina Institution: Centro Privado de Psicoterapias Session: 3570 - PARALLEL SHORT ORAL POSTER SESSIONS: DIGITAL & AI PSYCHIATRY - Part III Date: Fri, 25.09.2026 Blinded expert supervisors found 89% of the supervision feedback generated by a Spanish-language systemic strategic AI assistant clinically appropriate and 91% suitable for delivery to the therapist. The study aims to evaluate the clinical quality and model fidelity of supervision feedback generated by a Spanish-language AI assistant developed in Argentina and grounded in systemic strategic therapy. In a single-center exploratory pilot design, 15 consecutive real-world supervision cases from January to March 2024, predating the use of AI, were submitted to the assistant (cases involving minors, psychopharmacology, and repeat supervision were excluded, and identifiers were removed). The assistant was described not as the general-purpose model itself, but as a custom configuration built on Claude Opus 5, with a systemic strategic system prompt and a curated systemic therapy knowledge base that included Argentine authors. Fifteen blinded expert supervisors each rated three cases, with three raters per case, yielding 45 assessments in total. Theoretical appropriateness, clinical applicability, safety and risk, model adherence, and usefulness were rated on an anchored scale from 1 to 5; an overall score, a binary appropriateness judgment, and source attribution were also requested. Eighty-nine percent of assessments received a score of 4 or higher, indicating appropriateness (95% confidence interval, 77–95), and 91% were judged suitable for delivery to the therapist. Median scores were 4 for appropriateness, applicability, and overall quality, and 5 for safety, adherence, and usefulness; the interquartile range was 4–5 across all dimensions. Interrater agreement measured by the intraclass correlation coefficient was low, ranging from 0.19 to 0.33, which the authors interpreted as a consequence of ceiling effects rather than disagreement; rWG values ranged from 0.74 to 0.83, and the proportion of rater pairs whose scores differed by no more than one point ranged from 0.84 to 0.93. As a secondary finding, although all feedback was AI-generated, raters classified its source as AI in 63% of assessments (p = 0.13), and feedback attributed to humans received significantly higher overall quality scores (Mann–Whitney p < 0.001; rank-biserial r = 0.58). The authors interpret the results as supporting the feasibility of a purpose-built Spanish-language assistant as an aid to supervisors and laying the groundwork for a controlled equivalence study; limitations include the single-center design, ceiling effects, rater severity, the absence of statistical modeling of clustering, and an unvalidated assessment rubric. [View the original poster](https://cslide.ctimeetingtech.com/wcp26/attendee/eposter/file/1433) How to cite this study APA 7: Grosman, S. A. (2026, September 25). Expert-rated quality of artificial intelligence-assisted clinical supervision in strategic psychotherapy. an ongoing pilot study (id 5619) [Poster presentation]. 26th WPA World Congress of Psychiatry (WCP 2026), Stockholm, Sweden. https://cslide.ctimeetingtech.com/wcp26/attendee/eposter/file/1433 Vancouver: Grosman SA. Expert-rated quality of artificial intelligence-assisted clinical supervision in strategic psychotherapy. an ongoing pilot study (id 5619) [poster]. 26th WPA World Congress of Psychiatry (WCP 2026); 2026 Sep 23-26; Stockholm, Sweden. Available from: https://cslide.ctimeetingtech.com/wcp26/attendee/eposter/file/1433 Author names and title formatting were generated automatically from the congress system; please check them against the original poster before use. ## On the same topic - [Siber tanıkların davranışını belirleyen etkenler](https://alisanburak.com/en/wcp-2026-posterler/ev0041) - [Z kuşağında sosyal medyanın "lunapark etkisi"nin modellenmesi: örtük değişken yaklaşımı](https://alisanburak.com/en/wcp-2026-posterler/ev0046) - [DEHB'de dijital girişimler: etkililik ve çok modlu tedavi yaklaşımlarına entegrasyon](https://alisanburak.com/en/wcp-2026-posterler/ev0072) - [Algoritmaların inşa ettiği beyinler: dijital çağda DEHB, patolojiyi mi tanılıyoruz yoksa yeni bir normali mi patolojikleştiriyoruz?](https://alisanburak.com/en/wcp-2026-posterler/ev0077) - [Genç erişkin bir örneklemde nomofobi ile yaygın anksiyete bozukluğu arasındaki ilişki](https://alisanburak.com/en/wcp-2026-posterler/ev0111) [All WCP 2026 summaries](https://alisanburak.com/en/wcp-2026-posterler) --- Kaynak / Source: https://alisanburak.com/en/wcp-2026-posterler/5619 İletişim / Contact: Bağdat Caddesi, Veli Güneysu Apt. No:453 K:4 D:9, Suadiye / Kadıköy, İstanbul · +90 532 642 95 25 · info@alisanburak.com · https://alisanburak.com/bagdat-caddesi-psikoterapi-iletisim