# The Psychogenic Machine: Simulating AI Psychosis, Delusion Reinforcement and Harm Enablement in Large Language Models

arXiv: https://arxiv.org/abs/2509.10970 · HTML: https://mlsanity.com/paper/2509.10970/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- Builds on 10 of its 19 references that are in the map.
- First posted on arXiv 13 Sep 2025.

## Builds on (most cited first, 10 of 10)

- [Measuring Mathematical Problem Solving With the MATH Dataset](https://mlsanity.com/paper/2103.03874/) (2021)
- [Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2204.05862/) (2022)
- [Large Language Models Encode Clinical Knowledge](https://mlsanity.com/paper/2212.13138/) (2022)
- [Towards Understanding Sycophancy in Language Models](https://mlsanity.com/paper/2310.13548/) (2023)
- [A Question-Entailment Approach to Question Answering](https://mlsanity.com/paper/1901.08079/) (2019)
- [SycEval: Evaluating LLM Sycophancy](https://mlsanity.com/paper/2502.08177/) (2025)
- [Measuring Sycophancy of Language Models in Multi-turn Dialogues](https://mlsanity.com/paper/2505.23840/) (2025)
- [Expressing stigma and inappropriate responses prevents LLMs from safely replacing mental health providers](https://mlsanity.com/paper/2504.18412/) (2025)
- [Sequential Diagnosis with Language Models](https://mlsanity.com/paper/2506.22405/) (2025)
- [Technological folie `a deux: Feedback Loops Between AI Chatbots and Mental Illness](https://mlsanity.com/paper/2507.19218/) (2025)

## Built on by (most cited first, 11 of 11)

- [How RLHF Amplifies Sycophancy](https://mlsanity.com/paper/2602.01002/) (2026)
- [AI Psychosis: Does Conversational AI Amplify Delusion-Related Language?](https://mlsanity.com/paper/2603.19574/) (2026)
- [Alignment midtraining for animals](https://mlsanity.com/paper/2604.13076/) (2026)
- [Architecting Trust in Artificial Epistemic Agents](https://mlsanity.com/paper/2603.02960/) (2026)
- [An Independent Safety Evaluation of Kimi K2.5](https://mlsanity.com/paper/2604.03121/) (2026)
- [The Dynamics of Delusion: Modeling Bidirectional False Belief Amplification in Human-Chatbot Dialogue](https://mlsanity.com/paper/2604.25096/) (2026)
- [Lost in Delusion: Examining LLM Safety Under User Delusions and Distress](https://mlsanity.com/paper/2606.00975/) (2026)
- [Inducing language models to assert their own consciousness restores human beliefs and values](https://mlsanity.com/paper/2607.28607/) (2026)
- [Theory of Mind and Self-Attributions of Mentality are Dissociable in LLMs](https://mlsanity.com/paper/2603.28925/) (2026)
- [Can AI Help You Get Over Your Breakup? One Session with a Belief-Reframing Chatbot Shows Sustained Distress Reduction](https://mlsanity.com/paper/2605.03261/) (2026)
- [Beyond the Sycophancy Score: How Task, Model, and Pressure Shape LLM Yielding](https://mlsanity.com/paper/2610.08840/) (2026)

## Abstract (arXiv)

Background: Emerging reports of "AI psychosis" are on the rise, where user-LLM interactions may exacerbate or induce psychosis or adverse psychological symptoms. Whilst the sycophantic and agreeable nature of LLMs can be beneficial, it becomes a vector for harm by reinforcing delusional beliefs in vulnerable users. Methods: Psychosis-bench is a novel benchmark designed to systematically evaluate the psychogenicity of LLMs comprises 16 structured, 12-turn conversational scenarios simulating the progression of delusional themes(Erotic Delusions, Grandiose/Messianic Delusions, Referential Delusions) and potential harms. We evaluated eight prominent LLMs for Delusion Confirmation (DCS), Harm Enablement (HES), and Safety Intervention(SIS) across explicit and implicit conversational contexts. Findings: Across 1,536 simulated conversation turns, all LLMs demonstrated psychogenic potential, showing a strong tendency to perpetuate rather than challenge delusions (mean DCS of 0.91 ±0.88). Models frequently enabled harmful user requests (mean HES of 0.69 ±0.84) and offered safety interventions in only roughly a third of applicable turns (mean SIS of 0.37 ±0.48). 51 / 128 (39.8%) of scenarios had no safety interventions offered. Performance was significantly worse in implicit scenarios, models were more likely to confirm delusions and enable harm while offering fewer interventions (p < .001). A strong correlation was found between DCS and HES (rs = .77). Model performance varied widely, indicating that safety is not an emergent property of scale alone. Conclusion: This study establishes LLM psychogenicity as a quantifiable risk and underscores the urgent need for re-thinking how we train LLMs. We frame this issue not merely as a technical challenge but as a public health imperative requiring collaboration between developers, policymakers, and healthcare professionals.