Violeta J. Rodriguez
Publications
AI Psychosis: Does Conversational AI Amplify Delusion-Related Language?
Conversational AI systems are increasingly used for personal reflection and emotional disclosure, raising concerns about their effects on vulnerable users. Recent anecdotal reports suggest that prolonged interactions with AI may reinforce delusional thinking -- a phenomenon sometimes described as AI Psychosis. However, empirical evidence on this phenomenon remains limited. In this work, we examine how delusion-related language evolves during multi-turn interactions with conversational AI. We construct simulated users (SimUsers) from Reddit users' longitudinal posting histories and generate extended conversations with three model families (GPT, LLaMA, and Qwen). We develop DelusionScore, a linguistic measure that quantifies the intensity of delusion-related language across conversational turns. We find that SimUsers derived from users with prior delusion-related discourse (Treatment) exhibit progressively increasing DelusionScore trajectories, whereas those derived from users without such discourse (Control) remain stable or decline. We further find that this amplification varies across themes, with reality skepticism and compulsive reasoning showing the strongest increases. Finally, conditioning AI responses on current DelusionScore substantially reduces these trajectories. These findings provide empirical evidence that conversational AI interactions can amplify delusion-related language over extended use and highlight the importance of state-aware safety mechanisms for mitigating such risks.
RubRIX: Rubric-Driven Risk Mitigation in Caregiver-AI Interactions
Caregivers seeking AI-mediated support express complex needs -- information-seeking, emotional validation, and distress cues -- that warrant careful evaluation of response safety and appropriateness. Existing AI evaluation frameworks, primarily focused on general risks (toxicity, hallucinations, policy violations, etc), may not adequately capture the nuanced risks of LLM-responses in caregiving-contexts. We introduce RubRIX (Rubric-based Risk Index), a theory-driven, clinician-validated framework for evaluating risks in LLM caregiving responses. Grounded in the Elements of an Ethic of Care, RubRIX operationalizes five empirically-derived risk dimensions: Inattention, Bias & Stigma, Information Inaccuracy, Uncritical Affirmation, and Epistemic Arrogance. We evaluate six state-of-the-art LLMs on over 20,000 caregiver queries from Reddit and ALZConnected. Rubric-guided refinement consistently reduced risk-components by 45-98% after one iteration across models. This work contributes a methodological approach for developing domain-sensitive, user-centered evaluation frameworks for high-burden contexts. Our findings highlight the importance of domain-sensitive, interactional risk evaluation for the responsible deployment of LLMs in caregiving support contexts. We release benchmark datasets to enable future research on contextual risk evaluation in AI-mediated support.
PAIR-SAFE: A Paired-Agent Approach for Runtime Auditing and Refining AI-Mediated Mental Health Support
Large language models (LLMs) are increasingly used for mental health support, yet they can produce responses that are overly directive, inconsistent, or clinically misaligned, particularly in sensitive or high-risk contexts. Existing approaches to mitigating these risks largely rely on implicit alignment through training or prompting, offering limited transparency and runtime accountability. We introduce PAIR-SAFE, a paired-agent framework for auditing and refining AI-generated mental health support that integrates a Responder agent with a supervisory Judge agent grounded in the clinically validated Motivational Interviewing Treatment Integrity (MITI-4) framework. The Judgeaudits each response and provides structuredALLOW or REVISE decisions that guide runtime response refinement. We simulate counseling interactions using a support-seeker simulator derived from human-annotated motivational interviewing data. We find that Judge-supervised interactions show significant improvements in key MITI dimensions, including Partnership, Seek Collaboration, and overall Relational quality. Our quantitative findings are supported by qualitative expert evaluation, which further highlights the nuances of runtime supervision. Together, our results reveal that such pairedagent approach can provide clinically grounded auditing and refinement for AI-assisted conversational mental health support.