# Measurement Without Validity: The Compounding Reliability Problem in Agentic AI Evaluation

arXiv: https://arxiv.org/abs/2608.00794 · HTML: https://mlsanity.com/paper/2608.00794/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in General LLM agent frameworks and benchmarks, one of 16 directions in LLM agents and tool use.
- Builds on 28 of its 48 references that are in the map.
- First posted on arXiv 1 Aug 2026.

## Placed in

- LLM agents and tool use / [General LLM agent frameworks and benchmarks](https://mlsanity.com/agents/general-llm-agent-frameworks-and-benchmarks/)

## Builds on (most cited first, 28 of 28)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [SWE-bench: Can Language Models Resolve Real-World GitHub Issues?](https://mlsanity.com/paper/2310.06770/) (2023)
- [WebArena: A Realistic Web Environment for Building Autonomous Agents](https://mlsanity.com/paper/2307.13854/) (2023)
- [τ-bench: A Benchmark for Tool-Agent-User Interaction in Real-World Domains](https://mlsanity.com/paper/2406.12045/) (2024)
- [OSWorld: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computer Environments](https://mlsanity.com/paper/2404.07972/) (2024)
- [Chatbot Arena: An Open Platform for Evaluating LLMs by Human Preference](https://mlsanity.com/paper/2403.04132/) (2024)
- [GAIA: a benchmark for General AI Assistants](https://mlsanity.com/paper/2311.12983/) (2023)
- [Model Cards for Model Reporting](https://mlsanity.com/paper/1810.03993/) (2018)
- [τ2-Bench: Evaluating Conversational Agents in a Dual-Control Environment](https://mlsanity.com/paper/2506.07982/) (2025)
- [Measurement and Fairness](https://mlsanity.com/paper/1912.05511/) (2019)
- [Self-Preference Bias in LLM-as-a-Judge](https://mlsanity.com/paper/2410.21819/) (2024)
- [Agent-as-a-Judge: Evaluate Agents with Agents](https://mlsanity.com/paper/2410.10934/) (2024)
- [Establishing Best Practices for Building Rigorous Agentic Benchmarks](https://mlsanity.com/paper/2507.02825/) (2025)
- [Lost in Simulation: LLM-Simulated Users are Unreliable Proxies for Human Users in Agentic Evaluations](https://mlsanity.com/paper/2601.17087/) (2026)
- [Play Favorites: A Statistical Method to Measure Self-Bias in LLM-as-a-Judge](https://mlsanity.com/paper/2508.06709/) (2025)
- [AgentProcessBench: Diagnosing Step-Level Process Quality in Tool-Using Agents](https://mlsanity.com/paper/2603.14465/) (2026)
- [ECBD: Evidence-Centered Benchmark Design for NLP](https://mlsanity.com/paper/2406.08723/) (2024)
- [Judge's Verdict: A Comprehensive Analysis of LLM Judge Capability Through Human Agreement](https://mlsanity.com/paper/2510.09738/) (2025)
- [Annotation alignment: Comparing LLM and human annotations of conversational safety](https://mlsanity.com/paper/2406.06369/) (2024)
- [Agent-as-a-Judge](https://mlsanity.com/paper/2601.05111/) (2026)
- [The Measurement Imbalance in Agentic AI Evaluation Undermines Industry Productivity Claims](https://mlsanity.com/paper/2506.02064/) (2025)
- [Counting on Consensus: Selecting the Right Inter-annotator Agreement Metric for NLP Annotation and Evaluation](https://mlsanity.com/paper/2603.06865/) (2026)
- [Dialectal Toxicity Detection: Evaluating LLM-as-a-Judge Consistency Across Language Varieties](https://mlsanity.com/paper/2411.10954/) (2024)
- [Auditing Automated Evaluation, Error Propagation, and Runtime Mitigation in Tool-Using Language Agents](https://mlsanity.com/paper/2604.16706/) (2026)
- [Toward Robust LLM-Based Judges: Taxonomic Bias Evaluation and Debiasing Optimization](https://mlsanity.com/paper/2603.08091/) (2026)
- [Bench-2-CoP: Can We Trust Benchmarking for EU AI Compliance?](https://mlsanity.com/paper/2508.05464/) (2025)
- [EvalCards: A Framework for Standardized Evaluation Reporting](https://mlsanity.com/paper/2511.21695/) (2025)
- [Expert Evaluation and the Limits of Human Feedback in Mental Health AI Safety Testing](https://mlsanity.com/paper/2601.18061/) (2026)

## Built on by (most cited first, 1 of 1)

- [Finding Blind Spots in AppWorld and WorkArena Task Verifiers](https://mlsanity.com/paper/2610.09142/) (2026)

## Abstract (arXiv)

Agentic AI evaluation pipelines produce benchmark scores that justify deployment decisions, safety certifications, and regulatory compliance claims. No formal framework has yet characterized how validity degrades across the stages of these pipelines. We present a three-layer compounding validity model, Vtotal ≤ V1 × V2 × V3, that captures multiplicative degradation across task generation (V1), human-simulator calibration (V2), and automated judgment (V3). Under empirically grounded estimates, a pipeline retaining 70% validity at each stage is at most 34% valid against the intended construct (range 0.17-0.54 across the empirical estimate bounds). We examine the model's predictions against a structured survey of 55 published agentic evaluation papers, finding that approximately 82% of papers in this purposive sample apply structurally mismatched, incomplete, or absent inter-rater reliability (IRR) metrics, a pattern consistent with systematic V3 collapse. We further identify empirical evidence of V1 failures (task validity flaws in 7 of 10 popular benchmarks) and V2 miscalibration (up to 9 percentage points inter-simulator variance, with systematic demographic disparities for non-Standard American English speakers). We derive eight prescriptions grounded in psychometric science and domain-stratified reliability thresholds (ICC ≥ 0.70; α ≥ 0.67/0.70/0.80 by consequence level) that practitioners and benchmark authors can apply immediately. The framework provides a tractable knowledge-based tool for diagnosing and correcting evaluation pipeline validity before deployment decisions are made.