AI ethics evidence index Search
Source-linked daily record
Evidence record 273 · automatically gathered

When Aggregate Alignment Misleads: Auditing Policy Repair Without Per-State Expert Actions

Agentic AI systems are increasingly used to edit, refine, and repair decision policies, but evaluating these edits is difficult when per-state expert action labels are unavailable. We study this problem in a hotel-pricing simulator where an agentic policy editor receives only region-level diagnostic feedback: summaries of how its price distribution differs from a benchmark policy across time, inventory, and market regions. The editor cannot observe benchmark actions, benchmark source code, rewar

Record details

Published: 3 July 2026
Source: arXiv
Category: Research
Topics: Regulation · Safety & alignment · Healthcare · Agents & autonomy
Retrieved: 14 July 2026

source-onlyevidence status

These records share source-supplied organisations, an exact publisher byline, automatic topics or regions. The reason is shown on every link; related does not mean supporting, agreeing with or verifying this record.

Same publisher byline · Shared topic: Safety & alignment

Market-Alignment Risk in Pricing Agents: Trace Diagnostics and Trace-Prior RL under Hidden Competitor State

arXiv · 7 May 2026

Shared topic: Regulation · Shared topic: Safety & alignment

Relevance as a Vulnerability: How Web Retrieval Degrades Safety Alignment in LLM Agents

arXiv · 28 May 2026

Shared topic: Regulation · Shared topic: Safety & alignment

Why Do AI Agents Break Rules? How Framing, Context, and Social Signals Shape Compliance

arXiv cs.CY · 14 August 2026

Same publisher byline · Shared topic: Regulation

When Outcome Looks Right But Discipline Fails: Trace-Based Evaluation Under Hidden Competitor State

arXiv · 18 May 2026

Shared topic: Regulation · Shared topic: Safety & alignment

Think Before You Act -- A Neurocognitive Governance Model for Autonomous AI Agents

arXiv · 28 April 2026

Shared topic: Regulation · Shared topic: Safety & alignment

Four-Axis Decision Alignment for Long-Horizon Enterprise AI Agents

arXiv · 21 April 2026

How to cite this record

ethics.ai (3 July 2026), “When Aggregate Alignment Misleads: Auditing Policy Repair Without Per-State Expert Actions,” evidence record 273, https://ethics.ai/record/273 (originally published by arXiv).

JSON

Use and limitations

This page is a stable index and citation surface for a source record. ethics.ai did not author the underlying report and has not independently verified every claim. Automatic topics may be imperfect. For consequential use, quote and cite the original publisher.