Leading healthcare organizations, including Boston Children’s and CVS, pledge adherence to ethical AI principles, prioritizing “Fair, Appropriate, Valid, Effective, and Safe” outcomes. Prompted by the Biden Administration’s executive order on AI safety, these 28 providers and payers commit to transparent AI use, risk management, and equitable health solutions. Aligning with HIMSS AI in Healthcare Forum, this commitment underscores the critical need for responsible AI integration in healthcare. Simultaneously, recent regulatory updates emphasize transparency in AI algorithms. These initiatives aim to ensure patient safety, equity, and improved healthcare experiences, marking a pivotal step toward ethical AI implementation in the healthcare sector.
The healthcare industry embraces ethical AI adoption, echoing the Biden Administration’s call for responsible AI practices. Notably, Boston Children’s, CVS, and others pledge 28 commitments to prioritize ethical AI principles—Fairness, Appropriateness, Validity, Effectiveness, and Safety (FAVES). This follows the administration’s push for AI safety and trustworthiness across sectors. The recent HIMSS AI in Healthcare Forum accentuates the industry’s commitment to transparent and equitable AI implementation. Concurrently, regulatory updates emphasize algorithm transparency in certified health IT. These concerted efforts signify a crucial leap towards ensuring patient well-being, fairness, and improved healthcare outcomes through responsible AI integration.
Shortly after the Biden Administration’s comprehensive executive order on AI was released, which emphasized “safe, secure, and trustworthy” AI across various sectors, the White House announced these commitments focused on AI transparency, risk management, and responsibility in healthcare. The executive order, published on October 30, contains specific healthcare-related provisions, tasking the U.S. Department of Health and Human Services with establishing a system to collect reports of “harms or unsafe healthcare practices.”
Aligned with the inaugural day of the HIMSS AI in Healthcare Forum in San Diego on December 14, the Biden Administration revealed these voluntary commitments from the private sector regarding healthcare AI safety and security.
The 28 providers and payers have vowed to adopt more transparent and trustworthy practices in the procurement and utilization of AI-based tools. Additionally, they’ve committed to developing their machine models responsibly. These commitments are meant to uphold the “FAVES” principles, ensuring that AI contributes to healthcare outcomes that are Fair, Appropriate, Valid, Effective, and Safe, according to National Economic Advisor Lael Brainard, Domestic Policy Advisor Neera Tanden, and Director of the Office of Science and Technology Policy Arati Prabhakar.
As part of the agreement, these healthcare organizations have promised to:
1. Notify patients and customers when they encounter substantially AI-generated content that hasn’t been reviewed or edited by humans.
2. Embrace and adhere to a risk management framework for monitoring and mitigating potential harms associated with AI-powered applications.
3. Explore innovative approaches to AI that advance health equity, widen access to care, reduce costs, improve outcomes, alleviate clinician burnout, and enhance the patient experience.
This initiative coincides with a significant week of developments in healthcare AI. The Office of the National Coordinator for Health IT released final regulations focused on algorithm transparency, emphasizing the need for certified health IT to provide users with essential information about predictive algorithms. Moreover, the HIMSS AI in Healthcare Forum in San Diego has brought together industry leaders to deliberate on the potentials and risks of AI in healthcare.
White House officials underscored the importance of these commitments, highlighting the necessity for rigorous testing, risk mitigation, and human oversight in AI-enabled tools used for clinical decisions. They view the private-sector commitments as a crucial step in a broader effort to advance AI for the health and well-being of Americans and encourage more organizations to join these commitments in the future.
