Use Risk-Based AI/ML Regulation: AstraZeneca
Responding to an FDA request for comments on total lifecycle considerations for generative artificial intelligence-enabled medical devices, AstraZeneca says it is important that the agency use a risk-based approach to the regulation of artificial intelligence/machine learning (AI/ML). “Risk-based frameworks are an important regulatory tool used to facilitate innovation by appropriately reserving the highest level of evidence and performance requirements for products that present the significant potential risk while leveraging regulatory flexibility for lower risk products to allow advancements in technology,” the company says.
AZ also says it applauds FDA for recent efforts to increase internal expertise and resources in the area of AI/ML, and encourages it to ensure that internal expertise and development of agency thinking and policy extends beyond CDRH since CDER and CBER are increasingly responsible for the regulation of AI/ML technologies in the areas of drug endpoints, digital health technologies for remote data acquisition, combination products with medical device software constituent parts, and promotional materials.
The firm’s other two comments deal with leveraging a range of communication strategies and engaging in external collaboration.