{
  "abstract": "We present BODHI (Balanced, Open-minded, Diagnostic, Humble, and Inquisitive), an engineering framework for curiosity driven and humble clinical decision support artificial intelligence (AI) systems. Despite growing capabilities, large language models (LLMs) often express inappropriate confidence, conflating statistical pattern recognition with genuine medical understanding. BODHI addresses this through a dual reflective architecture that: (1) decomposes epistemic uncertainty into task specific dimensions, and (2) constrains model responses using virtue based stance rules derived from a Virtue Activation Matrix. We validate the framework through controlled evaluation on 200 clinical vignettes from HealthBench Hard, assessing GPT-4o-mini and GPT-4.1-mini across 5 random seeds (2000 total observations). Statistical analysis included bootstrap resampling, paired t tests, and effect size computation. BODHI improved overall clinical response quality (GPT-4.1-mini: +16.6 pp, p<0.0001, Cohen's d=11.56; GPT-4o-mini: +2.2 pp, p<0.0001, Cohen's d=1.56) and achieved very large effect sizes on curiosity (context seeking rate: Cohen's d=16.38 and 19.54) and humility (hedging: d=5.80 for GPT-4.1-mini) metrics. Crucially, 97.3% of GPT-4.1-mini responses and 73.5% of GPT-4o-mini responses included appropriate clarifying questions, compared with 7.8% and 0.0% at baseline, demonstrating the framework's effectiveness in eliciting information gathering behaviour. Findings suggest LLMs can be reliably constrained to operate within epistemic boundaries when provided with structured uncertainty decomposition and virtue aligned response rules, offering a pathway towards safer clinical AI deployment",
  "authors": [
    {
      "affiliations": [
        "Institut du Cerveau, Sorbonne Universite, Paris, France",
        "Royal Victorian Eye and Ear Hospital Ronald Lowe Library, Melbourne, Victoria, Australia"
      ],
      "name": "Janan Arslan"
    },
    {
      "affiliations": [
        "The University of Melbourne Melbourne School of Engineering, Melbourne, Victoria, Australia"
      ],
      "name": "Kurt Benke"
    },
    {
      "affiliations": [
        "MIT, Cambridge, Massachusetts, USA"
      ],
      "name": "Sebastian Andres Cajas Ordoñez"
    },
    {
      "affiliations": [
        "MIT, Cambridge, Massachusetts, USA"
      ],
      "name": "Rowell Castro"
    },
    {
      "affiliations": [
        "Laboratory for Computational Physiology, Harvard-MIT Division of Health Sciences and Technology, Cambridge, Massachusetts, USA",
        "Division of Pulmonary Critical Care and Sleep Medicine, Beth Israel Deaconess Medical Center, Boston, Massachusetts, USA"
      ],
      "name": "Leo Anthony Celi"
    },
    {
      "affiliations": [
        "MIT, Cambridge, Massachusetts, USA",
        "Fundación Valle del Lili, Cali, Valle del Cauca, Colombia"
      ],
      "name": "Gustavo Adolfo Cruz Suarez"
    },
    {
      "affiliations": [
        "The University of Melbourne School of Computing and Information Systems, Melbourne, Victoria, Australia"
      ],
      "name": "Roben Delos Reyes"
    },
    {
      "affiliations": [
        "UCL, London, UK"
      ],
      "name": "Justin Engelmann"
    },
    {
      "affiliations": [
        "Cambridge University Hospitals NHS Foundation Trust, Cambridge, UK"
      ],
      "name": "Ari Ercole"
    },
    {
      "affiliations": [
        "MIT, Cambridge, Massachusetts, USA"
      ],
      "name": "Almog Hilel"
    },
    {
      "affiliations": [
        "Mbarara University of Science and Technology, Mbarara, Western Region, Uganda"
      ],
      "name": "Leo Kinyera"
    },
    {
      "affiliations": [
        "MIT, Cambridge, Massachusetts, USA",
        "King’s IoPPN, London, UK"
      ],
      "name": "Maximin Lange"
    },
    {
      "affiliations": [
        "University of Bergen, Bergen, Norway"
      ],
      "name": "Torleif Markussen Lunde"
    },
    {
      "affiliations": [
        "Technetium Engineering, LLC, Florida, Florida, USA"
      ],
      "name": "Mackenzie J Meni"
    },
    {
      "affiliations": [
        "MIT, Cambridge, Massachusetts, USA",
        "Fundación Valle del Lili, Cali, Valle del Cauca, Colombia"
      ],
      "name": "Felipe Ocampo Osorio"
    },
    {
      "affiliations": [
        "ETH Zurich, Zürich, Switzerland"
      ],
      "name": "Anna E Premo"
    },
    {
      "affiliations": [
        "University of Zurich, Zürich, Switzerland"
      ],
      "name": "Jana Sedlakova"
    },
    {
      "affiliations": [
        "MIT, Cambridge, Massachusetts, USA"
      ],
      "name": "Pritika Vig"
    }
  ],
  "title": "Engineering framework for curiosity-driven and humble AI in clinical decision support",
  "uid": "27846a14-0466-5e1d-b8a9-bc9814b138d8"
}