Saved in:
Bibliographic Details
Main Author: Lopez-Martinez, Daniel
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2406.16455
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913402677362688
author Lopez-Martinez, Daniel
author_facet Lopez-Martinez, Daniel
contents Generative AI (GenAI) models have demonstrated remarkable capabilities in a wide variety of medical tasks. However, as these models are trained using generalist datasets with very limited human oversight, they can learn uses of medical products that have not been adequately evaluated for safety and efficacy, nor approved by regulatory agencies. Given the scale at which GenAI may reach users, unvetted recommendations pose a public health risk. In this work, we propose an approach to identify potentially harmful product recommendations, and demonstrate it using a recent multimodal large language model.
format Preprint
id arxiv_https___arxiv_org_abs_2406_16455
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Guardrails for avoiding harmful medical product recommendations and off-label promotion in generative AI models
Lopez-Martinez, Daniel
Artificial Intelligence
Generative AI (GenAI) models have demonstrated remarkable capabilities in a wide variety of medical tasks. However, as these models are trained using generalist datasets with very limited human oversight, they can learn uses of medical products that have not been adequately evaluated for safety and efficacy, nor approved by regulatory agencies. Given the scale at which GenAI may reach users, unvetted recommendations pose a public health risk. In this work, we propose an approach to identify potentially harmful product recommendations, and demonstrate it using a recent multimodal large language model.
title Guardrails for avoiding harmful medical product recommendations and off-label promotion in generative AI models
topic Artificial Intelligence
url https://arxiv.org/abs/2406.16455