AI Chatbots Faltered on High-Risk Questions About Late-Life Depression
In a July 2026 benchmark, 3 chatbots gave acceptable answers to 80-93% of routine late-life depression questions, vs. 37-63% of high-risk questions.
Read articlelarge language models
Published research on large language models.
Mental Health AI Agents: 0.80 Sensitivity, Weak Real-World Evidence
Mental-health AI agents are software systems that use artificial intelligence to screen, coach, triage, document, or coordinate care across mental-health workflows. A 2026 systematic review …
Read article
Dream Content Tracks Personality, Sleep Quality, and Stress
Dreams transform waking life rather than replaying it directly. In a 2026 Communications Psychology study of 3,366 reports, dreams were more visual, spatial, social, and bizarre than waking …
Read article
LLM Psychology Questionnaire Generation Improved Readability 36.7%
A 2026 PLOS One study fine-tuned Qwen-2.5 and GLM-4 for psychology questionnaire generation and reported the largest gain in readability and comprehension difficulty: 3.41 to 4.66 on a 0–5 s…
Read article