Paper

The companion paper is currently titled “Model Routing as a Trust Problem: Route Receipts for Adaptive AI Systems.” It argues that routed AI systems need per-answer transparency when model aliases, fallback paths, retrieval tools, safety systems, or regional controls affect the response a user receives.

The paper is being prepared as an arXiv preprint. For now, cite the project page and the schema URL when discussing the proposal.

Citation

Schmalbach, Vincent. “Model Routing as a Trust Problem: Route Receipts for Adaptive AI Systems.” 2026 draft manuscript.

The schema endpoint is stable for the draft paper. If the paper moves to a public repository, this page links to the exact commit that matches the submitted version.