THE SIGNAL
On January 6, 2025, the FDA issued draft guidance titled Considerations for the Use of Artificial Intelligence to Support Regulatory Decision-Making for Drug and Biological Products. The agency called it its first guidance on AI in the development of drugs and biological products.
The draft sets out a risk-based framework for establishing the credibility of an AI model for a particular context of use, which the FDA defines as how a model is used to address a certain question of interest. The agency said it has seen more than 500 drug and biological product submissions with AI components since 2016, and it encourages sponsors to engage early. The guidance was issued as a draft for public comment, and it remains a draft as of October 2026.
BEFORE
Hypothetical: a clinical development team uses a model to help flag which trial participants may need more frequent safety monitoring. Data science has documented performance. The question of interest and the context of use are written.
Whether that evidence is enough for this use waits on the head of regulatory affairs and the head of clinical safety, who are both preparing for an advisory meeting. The file is ready. The people who can release it are not.
WHAT THE WAIT COSTS
The protocol amendment stalls. Sites wait for instructions. The window for early engagement with the agency slips by another month.
AFTER
The credibility plan arrives as one decision packet: the question of interest, the context of use, how much the model's output will influence the decision, what happens if it is wrong, and the evidence so far.
Each owner signs their part with comments. By Friday the team knows whether to proceed, gather more evidence, or request early engagement with the FDA.
HOW WISDOMTWIN FITS
WisdomTwin.ai is building a sovereign judgment layer for regulated enterprises in banking, healthcare, legal, government, pharma and defense. It is designed to run inside your own environment, assemble the decision packet with the evidence attached, route it to the named person who holds the authority, and keep a record of who released the decision and why.
We don't replace judgment. We remove the wait.
TRY THIS THIS WEEK
List every model whose output feeds a regulatory decision.
Write each model's context of use in one sentence.
Name the person who accepts the credibility risk for each one.
SOURCES
U.S. Food and Drug Administration, FDA Proposes Framework to Advance Credibility of AI Models Used for Drug and Biological Product Submissions, January 6, 2025: https://www.fda.gov/news-events/press-announcements/fda-proposes-framework-advance-credibility-ai-models-used-drug-and-biological-product-submissions
Examples in this issue are hypothetical, and nothing here describes a customer result.
Book a 20-minute call: https://calendly.com/romanbodnarchuk/20min
Roman Bodnarchuk, Co-Founder and CEO, WisdomTwin.ai