Artificial Intelligence in Surgery Raises Major Legal and Ethical Challenges

Artificial intelligence is gradually transforming surgery by improving operation planning, real-time assistance, and even certain autonomous tasks. However, its adoption introduces technological, human, legal, and ethical risks that are still poorly addressed by current regulations.

Artificial intelligence systems in surgery can make diagnostic or treatment errors, compromise patients’ informed consent, or violate their privacy. These risks stem from three main sources: the tool’s inherent limitations, human errors related to its use, and organizational failures during its deployment. For example, a poorly trained algorithm can amplify biases present in the data, leading to unfair recommendations for certain patient groups. Similarly, a surgeon might over-rely on a plausible but incorrect suggestion from the tool, thereby abandoning their own clinical judgment.

Current legal frameworks, such as those in the European Union or the United States, attempt to regulate these technologies by classifying systems according to their level of risk. Requirements then vary in terms of testing, data management, and cybersecurity. However, these regulations struggle to keep pace with innovations, particularly in the face of models capable of continuous learning and adaptation. Protecting patient data remains a critical issue. In Europe, the General Data Protection Regulation imposes strict rules on the use of medical information, while in the United States, gaps in current laws expose patients to risks of re-identification of their data, even when anonymized.

Informed consent becomes more complex with the introduction of artificial intelligence. Patients, often unfamiliar with these technologies, may develop unrealistic expectations or, conversely, unfounded fears. Surgeons must therefore clearly explain the tool’s role: is it a diagnostic aid, a navigation guide, or an active assistant during the operation? They must also describe the safeguards in place, such as their ability to verify and override the tool’s recommendations. Additionally, surgeons must inform patients about their own training and experience with these systems.

The question of liability in the event of medical error becomes more complicated with the use of artificial intelligence. Traditionally, the surgeon assumes full responsibility for decisions made in the operating room. But if a system provides an erroneous recommendation that causes harm, who is responsible? The surgeon for following it, the developer for a design flaw, or the institution for negligent implementation? Courts could hold manufacturers liable if a defective or poorly tested algorithm led to an error. Hospitals, on the other hand, could be deemed negligent if they use an unvalidated or poorly maintained system. As for the surgeon, they remain bound to respect standards of care and cannot absolve themselves of responsibility by blaming the tool.

The risks are not limited to medical errors. Misuse of artificial intelligence can also erode trust between the patient and the surgeon. If the patient perceives that decisions are being made by an algorithm rather than by their doctor, the trust relationship may be weakened. Furthermore, a lack of transparency or an error by the tool can permanently damage this trust, both in the clinician and in the healthcare system as a whole.

To mitigate these risks, robust governance is essential. Healthcare institutions must establish dedicated committees to evaluate artificial intelligence systems before their adoption. These committees must ensure that developers provide comprehensive user manuals, training sessions, and safety guidelines. Continuous performance monitoring is also necessary to detect any degradation or drift in the model. Professional societies and medical institutions also have a role to play in establishing standards for the training and certification of surgeons using these tools.

Finally, collaboration between hospitals, developers, and regulators is essential to harmonize practices and ensure the safe and ethical use of artificial intelligence in surgery. Without this, technological progress could come at the expense of patient safety and quality of care.

Content References

Official Reference

DOI: https://doi.org/10.1007/s00464-026-12881-8

Title: Risk and liability in the deployment of AI systems for surgery: a SAGES white paper

Journal: Surgical Endoscopy

Publisher: Springer Science and Business Media LLC

Authors: Daniel A. Hashimoto; Jayson S. Marwaha; Sharon A. Lee; Steven Schwaitzberg; Mindy N. Duffourc