Research before turning claims into product.
CABI 2036 structures Astrynn’s public research around Institutional Agency Integrity. Public material focuses on the problem, evidence and falsifiable results; proprietary mechanisms remain private.
AI systems can cross changes in context, responsibility and conditions before producing real consequences. CABI 2036 studies how to preserve a rigorous reading of that reality without conflating technical permission, institutional authority and evidence.
The public version explains the problem space and high-level relationships. Proprietary mechanisms, internal models, schemas and decision logic are not published.
It provides a common discipline for investigating consequential agency problems, separating claims and requiring evidence before turning a hypothesis into product.