Blog
Tips and insights on mental health and emotional well-being
Burnout: the 3 ICD-11 dimensions and what helps
Burnout in ICD-11: exhaustion, cynicism and reduced efficacy at work. Recognize the signs, review your workload and distinguish self-help from burnout evidence.
How to cope with stress: techniques by timeframe
Evidence-based self-help techniques for stress — from grounding in a minute to habits over a week — and when stress is chronic and it's time to see a professional.
AI vs. Therapist: A Role-by-Role Map of What 2024–2025 Evidence Actually Shows
A therapist plays four roles. AI in 2024–2025 reaches near-human performance on two (technique delivery, parts of alliance), is triage-only on a third (clinical judgment), and cannot own the fourth (case-level diagnosis).
Five CBT chatbots compared by design and research evidence
SuDoSys, CRBot, Socrates 2.0, behavioral activation and GPT-4 problem solving: what each study measured, from simulated dialogue to user experience.
Digital CBT-I and AI chatbots for insomnia: what the studies measured
29 randomized trials support automated digital CBT-I. A separate eight-model experiment tests LLM dialogue quality. Here is what each result means.
Therapeutic Alliance With an AI Therapist: 527-User Study
WAI-SR score 3.76/5 with an AI chatbot — near in-person therapy levels. Who bonds with AI, where the alliance breaks, and what actually builds trust.
How AI studies suicide risk in text and medical records
Language markers, medical-record prediction and false positives answer different questions. A closer look at the evidence behind AI suicide-risk detection.
MIND-SAFE and the questions to ask before choosing an AI assistant
Boit and Patil propose MIND-SAFE as a design framework for mental-health chatbots. Its architecture, suggested metrics and practical questions for clinics.
Prompt engineering and testing for mental health chatbots
How MIND-SAFE and SuDoSys structure chatbot conversations, what the Li review actually reported, and what EmoAgent measured in simulations.
Rule-based chatbots and LLMs for depression: what the 2025 review found
Du et al. found a small depression benefit for rule-based chatbots, but no significant difference between chatbot types. Results and uncertainty explained.
DSM5AgentFlow and AI diagnosis on a synthetic benchmark
How DSM5AgentFlow links simulated interviews to DSM-5 criteria, what accuracy, precision and class F1 mean, and how the authors assessed explanations.
Guardrails for mental-health AI: what EmoAgent tested
EmoAgent tested character chatbots with simulated users. How EmoEval measures score changes and what the EmoGuard experiments found.
Do AI Therapists Work? Meta-Analysis of 35 Studies
35 studies, 17,000+ participants: AI chatbots measurably reduce depression and distress. The effect sizes, the limits, and who benefits most.
Multi-agent AI vs chatbot: what the MIND study found
What the 42% ablation result means in MIND, which content metrics were rated, and how its eight-student experiment assigned only two people to MIND.
Self-compassion and AI inner dialogue: a closer look at the MIND prototype
MIND explores role reversal and AI inner dialogue. Its 13% result concerns interaction ratings; only two participants used MIND in the mood experiment.
A Therapist in Your Pocket: Why Running an AI Therapist Directly on Your Phone Matters
An AI therapist that works offline and never sends data to servers. How on-device models protect the most sensitive data you produce — and what regulators require.
What MoPHES achieved on a counseling benchmark
MoPHES beat GPT-4.1 on ROUGE-1 in one benchmark setup, but ranked below GPT-4.1 and PsyChat on automated response ratings. What the comparison shows.
AI Ethics in Psychotherapy: Who's Responsible When an Algorithm Harms a Patient?
Who's liable when an AI therapist causes harm? We examine real incidents, FDA regulation, the EU AI Act, and what every user should know.
Therabot trial: what a 51% reduction in depression symptoms means
Therabot users had a 51% reduction in mean PHQ-9 scores by week eight. Read the trial results, comparison group and limits of that finding.
ChatGPT as a therapist: what the research actually tests
What ChatGPT can help with in self-reflection, what studies measured, and how memory and responses to distress have changed. Product facts checked October 2026.
Computational models of mental disorders and what they can explain
How computational psychiatry models reward, perception and symptoms, with examples from depression research and the limits of individual prediction.
Digital phenotyping: what phones and language can reveal about depression
How researchers study phone activity, voice and text as signals of mental health. Evidence from digital phenotyping and the meaning of screening metrics.
Just-in-Time Interventions: How AI Is Learning to Help During a Crisis
What if an app could support you at the exact moment you need it — automatically? We explain the JITAI concept and how AI is learning to respond to stress in real time.