New model, old risks: sociodemographic bias and adversarial hallucinations vulnerability in GPT-5.
cross_sectional · Level IV
Where this comes from
- Record sourced from PubMed, PMID 41935214.
- Also identified by DOI 10.1038/s41746-026-02584-8 and PMC identifier 13050324.
- Licence recorded as CC BY-NC-ND.
- Because redistribution is not established, this page shows the abstract only. Follow the links below for the full text.
Abstract
We re-evaluated GPT-5 using our published pipelines: 500 emergency vignettes across 32 sociodemographic labels for bias, and adversarial prompts with fabricated details. GPT-5 showed no measurable improvement over GPT-4o in sociodemographic-linked decision variation, with several LGBTQIA+ groups flagged for mental-health screening in 100% of cases. Adversarial hallucination rates were higher (65% vs 53% for GPT-4o); a mitigation prompt reduced this to 7.67%.