{
  "abstract": "ObjectivesTo determine whether single-layer evaluation characterises clinical risk in the final note of a multilingual ambient AI scribe, and where serious errors arise.\n\nDesignTwo-arm evaluation on one common clinical-risk scale, using a frozen, reference-aligned synthetic corpus.\n\nSettingScripted consultations spanning 77 languages and five complexity levels, from simple general practice to expert multidisciplinary-team handover.\n\nMain outcome measuresPer-session incidence of at least one serious (HIGH or CRITICAL) note-layer error in intrinsic (reference script to note, n=385) and end-to-end (automatic speech recognition (ASR) transcript to note, n=2,302) generation; three LLM raters classified discrepancies using a four-mode taxonomy.\n\nResultsIntrinsic note generation had a 3.4% serious-error rate, with no detectable gradient across complexity (1% to 6%; Cochran-Armitage p=0.42) or language resource (high 2%, medium 5%, low 4%). End-to-end serious errors rose steeply with complexity (L1 1% to L4/L5 24%) and language scarcity (high 7%, medium 10%, low 15%; OR 1.54 per tier, 95% CI 1.15-2.07; p=0.004, GEE clustered on language). Of serious in-note errors, 87% were ASR-derived and 7% note-originated; amplification exceeded correction threefold (fate entropy 1.19 bits). Repeated generation showed moderate reproducibility (Fleiss kappa 0.42); word error rate explained 24% of cross-language variance versus [~]1% for transcription risk density.\n\nConclusionsLow serious-error rates in individual layers did not preclude higher end-to-end note risk. Evaluation should therefore include the clinician-facing note, because component-level metrics cannot fully capture risk created or transformed at the transcription-to-generation interface.\n\nWhat is already known on this topicO_LIAmbient AI scribes are being deployed globally, and safety evaluation has characterised the transcription layer in isolation, most commonly with word error rate C_LIO_LIFrequency-based transcription metrics such as word error rate do not directly encode clinical consequence, and component-level performance may not represent the risk of the final generated note C_LIO_LILarge language models produce clinically relevant output stochastically, so identical prompts can yield materially different content C_LI\n\nWhat this study addsO_LINote generation from a perfect transcript had a low observed serious-error rate and no detectable complexity or language-resource gradient, whereas the end-to-end system showed steep gradients on both axes. C_LIO_LISerious note-layer errors were predominantly ASR-derived, but note generation transformed upstream errors variably, amplifying serious errors about three times more often than it corrected them C_LIO_LITranscription-layer metrics had limited predictive value for note-layer risk after accounting for complexity and language-level dependence; word error rate explained approximately one-quarter of cross-language variance C_LI\n\nHow this study might affect research, practice or policyO_LIEvaluation and procurement of ambient scribes should include end-to-end assessment of the clinician-facing note, with results stratified by consultation complexity and language where relevant C_LIO_LIA three-surface root-cause alignment using a four-mode root-cause taxonomy (note-originated, propagated-ASR, amplified-ASR and silent-correction) can attribute documentation errors to their point of origin or transformation C_LIO_LILanguage-related performance should remain a surveillance target because disparities can emerge after components are composed, even when corresponding gradients are not detectable in an individual layer C_LI",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": "PUBLISHAHEADOFPRINT",
  "arxiv": null,
  "author_total": 4,
  "authors": [
    {
      "name": "H. Bergman"
    },
    {
      "name": "V. Liu"
    },
    {
      "name": "B. Austin"
    },
    {
      "name": "R. Sanghera"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/7sva656t982st.bib",
  "cited_by": 0,
  "doi": "10.64898/2026.09.22.26363642",
  "fulltext_url": "https://www.medrxiv.org/content/10.64898/2026.09.22.26363642v1.full",
  "id": "7sva656t982st",
  "issns": [],
  "landing_url": "https://www.medrxiv.org/content/10.64898/2026.09.22.26363642v1",
  "language": null,
  "licence": "CC-BY-NC-ND",
  "licence_label": "CC BY-NC-ND",
  "licence_source": "article",
  "licence_url": "https://creativecommons.org/licenses/by-nc-nd/4.0/",
  "metadata_licence": null,
  "native_subjects": [
    [
      "medrxiv",
      "health informatics"
    ],
    [
      "openalex",
      "2718|Artificial Intelligence in Healthcare and Education"
    ]
  ],
  "openalex": "W7214083531",
  "pdf_url": "https://www.medrxiv.org/content/10.64898/2026.09.22.26363642v1.full.pdf",
  "pmcid": null,
  "pmid": null,
  "published": "2026-09-22",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/7sva656t982st.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:19:14Z",
      "native_id": "10.64898/2026.09.22.26363642",
      "source": "medrxiv",
      "url": "https://www.medrxiv.org/content/10.64898/2026.09.22.26363642v1"
    }
  ],
  "subjects": [
    {
      "field": "medicine",
      "origin": "medrxiv:health informatics",
      "subfield": "epidemiology-public-health"
    }
  ],
  "title": "Mind the gap: emergent clinical risk at the interface of two individually safe AI systems in a multilingual ambient scribe",
  "title_text": "Mind the gap: emergent clinical risk at the interface of two individually safe AI systems in a multilingual ambient scribe",
  "type": "preprint",
  "updated_at": "2026-10-30T12:19:14Z",
  "url": "https://research.draconitic.com/paper/7sva656t982st",
  "venue": "medRxiv",
  "versions": []
}