Are Clinicians Ethically Obligated to Disclose Their Use of Medical Machine Learning Systems to Patients? - Pre-print Article

Aug 5, 2026

Upload by Violet

Common ethical guidance holds that clinicians must disclose their use of medical machine learning systems to patients, and that nondisclosure is a moral fault requiring accountability. This paper challenges that view, called the “disclosure thesis.” It reviews four standard justifications—risk-based, rights-based, materiality, and autonomy arguments—and argues each is unconvincing, so the disclosure thesis should be rejected. Mandated disclosure may also harm patients by enabling stakeholders to evade responsibility for harms linked to improper implementation and use.

LanguageEnglish

CategoryHealthcare

FormatPDF

Pages21

Reading Time53 min

AccessFree to download

Introduction

The disclosure thesis and its rejection

Four main arguments

Types of medical ML systems

This is a pre-print version of an article published as: Hatherley, Joshua. 2024. Are clinicians ethically obligated to disclose their use of medical machine learning systems to patients? Journal of Medical Ethics, forthcoming. 10. 1136/jme-2024-109905 . Please cite the published version. ARE CLINICIANS ETHICALLY OBLIGATED TO DISCLOSE THEIR USE OF MEDICAL MACHINE LEARNING SYSTEMS TO PATIENTS? Joshua Hatherley Abstract: It is commonly accepted that clinicians are ethically obligated to disclose their use of medical machine learning systems to patients, and that failure to do so would amount to a moral fault for which clinicians ought to be held accountable. Call this “the disclosure thesis.” Four main arguments have been, or could be, given to support the disclosure thesis in the ethics literature: the risk-based argument, the rights-based argument, the materiality argument, and the autonomy argument. In this article, I argue that each of these four arguments are unconvincing, and therefore, that the disclosure thesis ought to be rejected. I suggest that mandating disclosure may also even risk harming patients by providing stakeholders with a way to avoid accountability for harm that results from improper applications or uses of these systems. 1. Introduction Machine learning (ML) systems are rapidly being developed to assist in a broad range of clinical tasks, including diagnosis, risk prediction, treatment recommendations, and patient monitoring. [1,2] The use of these systems in medicine holds significant potential to improve patient health, health equity, and medical efficiency. [3,4] However, the use of ML systems in medicine has also generated various ethical dilemmas and concerns. [5-8] Medical ML systems may exacerbate existing health inequities due to their vulnerability to algorithmic bias. [9] Additionally, the reasoning processes of these systems are often opaque, making it difficult for users to understand how they arrive at their outputs. [10] These concerns have prompted discussion about clinicians' ethical obligations in communicating with patients about their use of medical ML systems,[11,12] and particularly whether clinicians are ethically obligated to disclose their use of these systems to patients. [13-15] In the ethics and regulatory literatures, the received view appears to be that clinicians are ethically obligated to disclose their use of medical ML systems to patients, or that failure to do so would amount to a moral failing for which clinicians ought to be held accountable. [16- 19] . According to Frank Ursin and co-authors, for instance, failure to disclose the use of a medical ML system “may amount to a form of deceit”(pg. 2) . [18] I will refer to this claim that clinicians are ethically obligated to disclose their use of medical ML systems as “the disclosure thesis.” There are four main arguments that have been, or could be used to defend the disclosure thesis. First, the “risk-based argument”, which states that clinicians are ethically obligated to disclose their use of medical ML systems because these systems generate risks to patient safety that warrant disclosure (section 2) . Second, the “rights-based” argument, which states that clinicians are obligated to disclose because using a medical ML system without a patient’s knowledge violates one or more of their moral rights as patients (section 3) . Third, the “materiality” argument, which states that a clinician must disclose their use of a medical ML system because this information is “material” to a patient’s decision (section 4) . Fourth, the “autonomy” argument, which states that a clinician must disclose because using a medical ML system without a patient’s knowledge threatens shared decision-making and patient autonomy (section 5) . My aim in this article is to evaluate the disclosure thesis by providing a critical assessment of each of these supporting arguments. Rather than assessing whether it would be prudent or m

FAQs

What is the “disclosure thesis” discussed in the article?

The article defines the disclosure thesis as the accepted claim that clinicians are ethically obligated to disclose their use of medical machine learning systems to patients, and that failing to do so is a moral fault warranting accountability.

Which four arguments are evaluated to support the disclosure thesis?

The paper considers the risk-based argument, rights-based argument, materiality argument, and autonomy argument as the main justifications that could defend the disclosure thesis.

Why does the article argue that mandated disclosure could harm patients?

It suggests disclosure requirements may allow clinicians, hospitals, developers, and other stakeholders to avoid accountability for harms arising from improper design, use, or implementation, and may socially license inappropriate applications.

Aug 5, 2026

Upload by Violet