Skip to content

Can artificial intelligence provide effective feedback in online therapy for social anxiety?

PSY-CHAI: a six-arm double-blind randomized controlled non-inferiority trial with factorial deception design comparing ai-generated and human therapist feedback in internet-based cognitive behavioural therapy for adults with social anxiety disorder

Status
Recruiting
Phases
Unknown
Study type
Interventional
Source
ISRCTN
Registry ID
ISRCTN11343758
Enrollment
480
Registered
2026-04-09
Start date
2025-09-01
Completion date
Unknown
Last updated
2026-04-20

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

Social anxiety disorder Mental and Behavioural Disorders

Interventions

Participants are randomized to one of six parallel groups. Four active groups form a 2x2 factorial design: (1) AI-generated feedback, participants told AI (transparent AI, n=80)
(2) AI-generated feedback, participants told human therapist (deceptive AI, n=80)
(3) human therapist feedback, participants told human therapist (transparent human, n=80)
(4) human therapist feedback, participants told AI (deceptive human, n=80). Two control groups complete the design: (5) unguided self-help with full treatment content but no personalized feedback (n=8
(6) waitlist control (n=80). All active groups (1-5) receive the 9-week SOFIE internet-based cognitive behavioural therapy protocol delivered via the Iterapi treatment platform. The protocol is based

Sponsors

Stockholm University
Lead Sponsor

Eligibility

Sex/Gender
All
Age
18 Years to 120 Years

Inclusion criteria

Inclusion criteria: 1. Age 18 years or older 2. Primary diagnosis of social anxiety disorder according to DSM-5 criteria, assessed via structured self-report screening 3. Score of 30 or above on the Liebowitz Social Anxiety Scale, Self-Report version (LSAS-SR), indicating at least moderate social anxiety 4. Stable psychotropic medication for at least 3 months, or no psychotropic medication 5. No concurrent psychological treatment 6. Ability to read and write Swedish 7. Access to the internet

Exclusion criteria

Exclusion criteria: 1. Severe depression as indicated by a score of 20 or above on the Patient Health Questionnaire-9 (PHQ-9) 2. Active suicidal ideation as indicated by PHQ-9 item 9 > 2 points at screening 3. Non-serious or invalid responses at clinical screening

Countries

Sweden

Contacts

Public ContactPer Carlbring
per.carlbring@psychology.su.se+468163920

Outcome results

None listed

Source: ISRCTN (via WHO ICTRP) · Data processed: Apr 23, 2026