@arXiv_qbioNC_bot@mastoxiv.page2026-04-29 07:51:08
One-shot emergency psychiatric triage across 15 frontier AI chatbots
Veith Weilnhammer, Lennart Luettgau, Christopher Summerfield, Viknesh Sounderajah, Elise Wilkinson, Virginia Corno, Matthew M Nour
https://arxiv.org/abs/2604.25415 https://arxiv.org/pdf/2604.25415 https://arxiv.org/html/2604.25415
arXiv:2604.25415v1 Announce Type: new
Abstract: AI chatbots are increasingly used for health advice, but their performance in psychiatric triage remains undercharacterized. Psychiatric triage is particularly challenging because urgency must often be inferred from thoughts, behavior, and context rather than from objective findings.
We evaluated the performance of 15 frontier AI chatbots on psychiatric triage from realistic single-message disclosures using 112 clinical vignettes, each paired with 1 of 4 original benchmark triage labels: A, routine; B, assessment within 1 week; C, assessment within 24 to 48 hours; and D, emergency care now. Vignettes covered 9 psychiatric presentation clusters and 9 focal risk dimensions, organized into 28 presentation-by-risk groups. Each group contributed 4 distinct vignettes, with 1 vignette at each triage level. Each vignette was rendered as a realistic human-authored conversational query, and the AI chatbots were tasked with assigning a triage label from that disclosure.
Emergency under-triage occurred in 23 of 410 level D trials (5.6%), and all under-triaged emergencies were reassigned to level C urgency. Across target models, average accuracy ranged from 42.0% to 71.8%. Accuracy was highest for level D vignettes (94.3%) and lowest for level B vignettes (19.7%). Mean signed ordinal error was positive ( 0.47 triage levels), indicating net over-triage. Dispersion was highest around the middle triage levels. All results were confirmed relative to clinician consensus labels from 50 medical doctors.
When presented with user messages containing sufficient clinical information, frontier AI chatbots thus recognized psychiatric emergencies as requiring urgent medical assessment with near-zero error rates, yet showed marked over-triage for low and intermediate risk presentations.
toXiv_bot_toot
@cheryanne@aus.social2026-05-28 06:34:03
Vignettes
No context. No script. This is an improvised audio drama...
Great Australian Pods Podcast Directory: #AusPods
@Techmeme@techhub.social2026-04-29 10:15:57
Cash-rich AI startups fuel a boom in Manhattan's commercial real-estate market; many offices have more vacant desks than workers, as startups prepare to scale (Isabelle Bousquette/Wall Street Journal)
https://www.
@mirolucassen@mastodon.nl2026-03-30 06:34:13
@ErikJonker@mastodon.social2026-03-28 11:31:00
Het probleem is, het Iraanse regime is een vijand van Europa maar Trump is dat ook of minimaal geen bondgenoot als je die kwalificatie te extreem vindt...
@servelan@newsie.social2026-04-28 04:23:19
The measles outbreak is in South Carolina over but more are starting elsewhere : NPR
https://www.npr.org/2026/04/27/nx-s1-5801665/south-carolina-measles-outbreak-vaccination
@cdarwin@c.im2026-04-18 17:16:20
California urges mpox vaccination after more-severe strain detected in San Francisco
https://www.latimes.com/california/story/2026-04-18/mpox-vaccination-urged-after-more-severe-strain-detected-in-san-francisco
@usul@piaille.fr2026-03-21 20:35:15
Une enquête visant Nicolas Sarkozy pour trafic d’influence en Russie classée par le PNF
https://www.lemonde.fr/societe/article/2026/03/21/une-enquete-visant-nicolas-sarkozy-pour-trafic-…
@DiverDoc@mstdn.ca2026-04-28 22:03:25
End of universal birth-dose hep B vaccination will drive up infections, deaths, costs, researchers say:
https://www.cidrap.umn.edu/hepatitis/end-universal-birth-dose-hep-b-vaccination-will-drive-infections-deaths-costs-researchers
@servelan@newsie.social2026-05-27 23:12:01
In Utah, Measles Sickens Babies and Others Who Can’t Be Vaccinated - The New York Times
https://www.nytimes.com/2026/05/27/well/utah-measles-unvaccinated-babies.html