Morning Edition · Wednesday, September 2, 2026Published at 2:20 AM EDT · New York

New Preprint Proposes Auditing Whether a Model API Actually Serves the Model It Advertises

AgentProv identifies a served backbone through its tool-use behavior rather than its text output, on the argument that providers may quantize or substitute models to cut serving costs.

A preprint posted to arXiv proposes AgentProv, a method for auditing commercial large language model (LLM) application programming interfaces by probing tool-use policy. The authors start from a commercial fact: an API advertises a specific…