How we’re using AI tools to improve psychedelic-drug research

2026-05-21 · Source: Machine learning : nature.com subject feeds · Field: Health & Wellbeing — Mental Health & Psychological Support, Health & Medical Research, Medical Devices & Health Technology · Depth: Intermediate, short

Summary

Cognitive scientist Félix Schoeller and Joshua White, founder of Fireside Project, developed Lucy, an artificial-intelligence-powered tool designed to improve training for facilitators of psychedelic-assisted therapy. Inspired by Fireside Project's database of thousands of anonymized support calls from people undergoing psychedelic experiences, Lucy addresses limitations of traditional role-play training. The platform uses large-language models (LLMs) to create diverse, realistic AI "clients" that simulate individuals in altered states, allowing trainees to practice skills for conditions like PTSD or depression with drugs such as psilocybin and MDMA. Lucy's web-based format aims to lower costs and scale training, which is crucial given the increasing demand for facilitators in US states like Oregon and Colorado. Currently, 400 clinicians, researchers, social workers, and educators are testing a pilot version, gaining experience with unusual cognitive and emotional presentations. The tool also helps standardize training for clinical trials and identifies effective interaction techniques, such as genuine human connection, by analyzing call data.

Key takeaway

For mental health organizations scaling psychedelic-assisted therapy programs, consider integrating AI simulation platforms like Lucy to standardize facilitator training. This approach allows your team to practice diverse, realistic scenarios, potentially lowering costs and accelerating the availability of qualified professionals. You should explore how such tools can provide objective feedback on therapist performance and identify effective interaction techniques, enhancing overall treatment quality and research consistency.

Key insights

AI-powered simulation tools can standardize and scale specialized therapy training using real-world interaction data.

Principles

• Real-world data enhances training realism.
• Standardized training improves research comparability.
• Genuine human connection is key to support.

Method

Analyze anonymized support call data, review existing training programs, design a scenario builder, and use LLMs to simulate client interactions and provide feedback.

In practice

• Simulate diverse client scenarios for practice.
• Tailor training to specific drugs or conditions.
• Identify therapist burnout indicators.

Topics

• Psychedelic-Assisted Therapy
• AI Training Simulation
• Large Language Models
• Mental Health Training
• Clinical Trials
• Fireside Project

Best for: NLP Engineer, AI Scientist, AI Product Manager, Domain Expert, Research Scientist, Policy Maker

Related on AIssential

• Mila’s AI drives world’s largest study on psychedelics — Psychedelics consistently flatten brain hierarchies, enabling communication between isolated regions.
• Good News & Bad News: AI is better than most therapy for some people. You need to understand some nuance, but its genuinely extraordinarily valuable. — AI provides uniquely personalized, accessible, and nuanced mental health support, often surpassing traditional therapy for specific needs.
• TherapyGym: Evaluating and Aligning Clinical Fidelity and Safety in Therapy Chatbots — TherapyGym evaluates and aligns therapy chatbots using clinical fidelity and safety metrics, improving performance through RL with expert-validated feedback.
• AI Improving Dose Decisions and Patient Outcomes in Oncology- with Shefali Kakar of Novartis — AI-driven data integration refines oncology dose decisions, safety, and patient variability, accelerating drug development.
• Two AI-based science assistants succeed with drug-retargeting tasks — AI-powered agentic systems can accelerate scientific discovery by synthesizing vast literature and generating testable hypotheses.
• Collaborating on a nationwide randomized study of AI in real-world virtual care — A nationwide randomized study will rigorously evaluate conversational AI in real-world virtual healthcare settings.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Machine learning : nature.com subject feeds.