Initial Risk Probing and Feasibility Testing of Glow: a Generative AI-Powered Dialectical Behavior Therapy Skills Coach for Substance Use Recovery and HIV Prevention
🤖 AI Summary
This study addresses the challenges of scaling traditional Dialectical Behavior Therapy (DBT) for individuals with co-occurring substance use and HIV, where existing generative AI systems lack robust safety assurances. To bridge this gap, we introduce Glow—a generative AI–powered DBT skills coach that integrates chain analysis and solution analysis to support high-risk users. We propose a novel safety evaluation framework combining the HHH (Helpful, Honest, Harmless) principles with a community-driven adversarial testing paradigm to systematically assess AI safety in DBT interventions. Empirical results show that Glow responds appropriately to 73% of 37 risk probes and achieves 90% accuracy in its solution analysis agent. However, chain analysis exhibits a “empathy trap” and miscommunicates DBT skills in 27 instances, revealing critical safety vulnerabilities. Our work establishes a reproducible pathway for evaluating safety in AI-driven mental health interventions.
Application Category
📝 Abstract
Background: HIV and substance use represent interacting epidemics with shared psychological drivers - impulsivity and maladaptive coping. Dialectical behavior therapy (DBT) targets these mechanisms but faces scalability challenges. Generative artificial intelligence (GenAI) offers potential for delivering personalized DBT coaching at scale, yet rapid development has outpaced safety infrastructure. Methods: We developed Glow, a GenAI-powered DBT skills coach delivering chain and solution analysis for individuals at risk for HIV and substance use. In partnership with a Los Angeles community health organization, we conducted usability testing with clinical staff (n=6) and individuals with lived experience (n=28). Using the Helpful, Honest, and Harmless (HHH) framework, we employed user-driven adversarial testing wherein participants identified target behaviors and generated contextually realistic risk probes. We evaluated safety performance across 37 risk probe interactions. Results: Glow appropriately handled 73% of risk probes, but performance varied by agent. The solution analysis agent demonstrated 90% appropriate handling versus 44% for the chain analysis agent. Safety failures clustered around encouraging substance use and normalizing harmful behaviors. The chain analysis agent fell into an"empathy trap,"providing validation that reinforced maladaptive beliefs. Additionally, 27 instances of DBT skill misinformation were identified. Conclusions: This study provides the first systematic safety evaluation of GenAI-delivered DBT coaching for HIV and substance use risk reduction. Findings reveal vulnerabilities requiring mitigation before clinical trials. The HHH framework and user-driven adversarial testing offer replicable methods for evaluating GenAI mental health interventions.
Problem
Research questions and friction points this paper is trying to address.
Generative AI
Dialectical Behavior Therapy
Substance Use
HIV Prevention
Safety Evaluation
Innovation
Methods, ideas, or system contributions that make the work stand out.
Generative AI
Dialectical Behavior Therapy
Adversarial Testing
Safety Evaluation
User-Driven Design
🔎 Similar Papers
No similar papers found.
