The Addition of Artificial Intelligence to Consent-as-Authority: Development of Care-Ethical Consent-as-Trust

Increased accessibility of specialized care in under-resourced medical scenarios is a frequently cited benefit of integration of artificial intelligence (AI) into medical care. Non-specialized healthcare workers or general practitioner physicians may use AI to enhance their medical skills and diagnose conditions or perform exams which they would not be able to perform without AI assistance. Under a care ethics framework, the use of medical AI to “upskill” providers presents a challenge to Tronto’s conception of consent-as-authority. Consent-as-authority both legitimizes and assigns responsibility within the doctor-patient relationship. Granting consent-as-authority necessitates that the patient make a judgement about the competency of the doctor and, if they consent, decide that the physician (and the broader medical team) has sufficient medical competency to care for them. However, when a physician is unable to perform the diagnostic skill or exam the patient needs without AI, this exercise is jeopardized. The question of medical competence seems to be unnecessary when the medical encounter is not dependent on the skills of the physician. In this work, we examine how consent-as-authority may be transformed into consent-as-trust, where the patient assigns responsibility and provides informed consent based only on the doctor’s competency as a caregiver. This new formulation provides an avenue by which the physician-patient relationship may adhere to the four principles of caring relationships when medical competency is circumvented by AI integration.