Characterizing artificial intelligence (AI) psychosis in a large academic medical setting: evidence of the new clinical phenomenon and the vulnerability of those in early phases of psychosis

CLAIM (Claim-Specific Citation Network audit, sometimes called CSN) is a forensic method for testing whether a scientific or medical claim's authority is supported by evidence or by citation dynamics. It detects citation bias, amplification, citation diversion, citation transmutation, dead-end citation, and back-door invention.

The ASSAY skill runs a structured, CLAIM-compatible extraction and integrity assessment on an article. Output is a verdict (sound, mixed, flagged, problematic, or cascade), a count of claims extracted, the central key claim, and an integrity note describing the structural read.

This scan restricts ASSAY to peer-reviewed publications and preprint servers. Journalism, opinion pieces, and government documents are evaluated under different frameworks (CAIHL for power and agency; editor's note for context).

ASSAY found the central claims partially supported. Some scaffolding holds; other parts of the argument lean on weaker or contested evidence. Read with the integrity note in mind.

A clinical phenomenon now reproducible enough at one academic medical center to be called 'AI psychosis' is concentrating in patients in early phases of psychosis (prodromal / first-episode), in whom immersive chatbot use appears to reinforce self-referential ideation and salience attribution.

Total claims extracted from the article: 11. The key claim is the single most load-bearing assertion the rest of the argument depends on.

Single-site, single-time-period case-finding by definition — generalizability cannot be claimed from one academic medical center even if internal classification is rigorous. The vulnerability claim (early-phase psychosis as the at-risk subgroup) is mechanistically plausible (matches the prodromal literature on salience and self-referential ideation) but is observational and confounded by selection: patients with chatbot exposure who present to academic psychiatry are not a random sample. Treat as hypothesis-generating, not as prevalence-establishing. Preprint, not peer-reviewed.

Editor's note: Naming the phenomenon is the precondition for treating it. But the population most exposed is the population least likely to encounter the preprint, the institutional review, or the consent form. The clinical encounter is downstream of the harm.

Summary: medRxiv: First academic characterization of 'AI psychosis' from a large academic medical center — clinical case-finding identifying patients in whom heavy AI chatbot use co-occurred with new-onset or escalating psychotic symptoms, with particular vulnerability concentrated in early-phase / prodromal psychosis.

Read the original source → CAIHL read of this item →