The Coalition for Health AI, formed in December 2022, has called on public and private partners via a JAMA study released Dec. 20, 2023, to develop assurance labs for the testing of artificial intelligence tools for healthcare based on arranged standards.
CHAI is concerned with the equity of current AI testing – with different standards and resources, inequitable distribution and application of these technologies is inevitable. By creating assurance labs as a "shared resource", there is an opportunity for "accelerating the pace of development and innovation, responsible and safe AI deployment, and successful market adoption," as the study.
Independent third-party testing lessens the possibility of bias, and provides a path for accountability and adherence to assurance standards. CHAI appreciates the need for local applications of all AI technologies, but maintains that the preferred method of testing is at a larger, more nation-wide scale.
According to CHAI, ongoing monitoring of AI performance is critical to maintaining credibility and accountability for AI models, and the best way to achieve constant monitoring is via assurance labs.