# The Delegation Blind Spot: Auditing Product Decisions from Agent Choices

arXiv: https://arxiv.org/abs/2609.26642 · HTML: https://mlsanity.com/paper/2609.26642/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 8 of its 13 references that are in the map.
- First posted on arXiv 22 Sep 2026.

## Builds on (most cited first, 8 of 8)

- [Detecting and Correcting for Label Shift with Black Box Predictors](https://mlsanity.com/paper/1802.03916/) (2018)
- [Prediction-Powered Inference](https://mlsanity.com/paper/2301.09633/) (2023)
- [Active Statistical Inference](https://mlsanity.com/paper/2403.03208/) (2024)
- [Linear Partial Monitoring for Sequential Decision-Making: Algorithms, Regret Bounds and Applications](https://mlsanity.com/paper/2302.03683/) (2023)
- [A Framework for Studying AI Agent Behavior: Evidence from Consumer Choice Experiments](https://mlsanity.com/paper/2509.25609/) (2025)
- [Partial Identifiability in Discrete Data With Measurement Error](https://mlsanity.com/paper/2012.12449/) (2020)
- [Robust Sampling for Active Statistical Inference](https://mlsanity.com/paper/2511.08991/) (2025)
- [A Revealed Preference Framework for AI Alignment](https://mlsanity.com/paper/2603.27868/) (2026)

## Abstract (arXiv)

Successful agent execution need not identify which future product improvement its user would value. We present a decision-specific audit that maps a declared observation channel and product-value contrast to compatible intervals and witness populations. Its foundations are established identification and decision theory; the contribution is an executable measurement workflow and a controlled study of its limits. A frozen experiment makes 4,800 requests to two pinned model snapshots on shared synthetic tasks. All 36 conservative primary intervals remain unresolved despite different execution accuracy. An exploratory 2,400-call follow-up records supplied preferences and resolves three of nine comparisons per model. A deterministic extractor resolves seven of nine without model calls or calibration observations, exposing unnecessary uncertainty introduced by model-generated reports. A further 14,400 controlled multinomial simulations distinguish structural ambiguity from weak identification and finite calibration precision. We propose a source-labeled decision receipt and provide an offline viewer for inspecting the audit. These results motivate preserving decision-relevant structured input and diagnosing why a decision is unresolved before collecting more telemetry. The study contains no human participants or real customer outcomes. Full proofs, raw model provenance, controlled experiments, and reproducible analyses accompany the report.