The traditional model of “informed consent” is ceasing to work in a world of big data and AI, where information is endlessly recombined and repurposed. Experts show that real power over data is now concentrated in the hands of platforms, while the promised autonomy of users turns into an illusion. Regulators are trying to intervene, but without new approaches to data governance, “free choice” will remain a ritual rather than a safeguard.

The U.S. Food and Drug Administration switched on its agency-wide generative AI system Elsa on June 2, 2025. The promise was faster scientific reviews, less bureaucratic “busywork,” and ultimately quicker access to new therapies. It’s only been six months, and the system is not without its flaws, but it’s definitely marked the start of a new AI era in how food, drugs, and devices are regulated. What effects of deployment can we see so far?