Evaluating AI Alignment in LLMs: Output Analysis of Value Priorities Across 75 Models with Human Benchmarking

TL;DR

This paper introduces a novel output-based method to evaluate AI alignment in LLMs by comparing their expressed value priorities with human judgments across 75 models, revealing both similarities and systematic differences.

Contribution

It develops a scalable profile-fidelity metric for auditing LLMs' value alignment with humans, highlighting the stability and variability of their value-priority structures.

Findings

LLM outputs show high stability within models and similar value-priority structures across models.

Most models replicate human value orderings but often exaggerate differences between priorities.

Profile fidelity varies widely and does not correlate strongly with model size or recency.

Abstract

Large language models (LLMs) are increasingly used in human-AI interaction research and practice, yet existing capability and safety benchmarks reveal little about the value priorities these systems express or how those priorities correspond to human judgements. Across three studies, we introduce an output-based approach to evaluating one facet of AI alignment by treating LLM-generated text as behavioural data and comparing expressed value-priority profiles with a human reference. Study 1 used inductive qualitative analysis to derive six themes of optimal AI functioning, namely Performance, Adaptive Capacity, Social Good, Ethics and Responsibility, Relational Integration, and Agency. Study 2 showed that LLM outputs were highly stable within models and converged on a common value-priority structure across models, indicating reliable and comparable value profiles. Study 3 benchmarked 75 contemporary LLMs against 376 human respondents using a profile-fidelity metric capturing both the relative ordering of priorities and the calibration of between-priority differences. Although most models reproduced the human ordering of values, some systematically exaggerated the differences between them, showing that models can appear aligned on conventional benchmarks while still diverging from human value calibration. Profile fidelity varied substantially across models and did not consistently scale with size, recency, or capability tier. Both LLMs and humans converged on a deprioritisation of Agency, raising important questions about the development of increasingly agentic AI systems. For research and applied use, the six themes and profile-based metric provide a scalable method for auditing LLM value profiles before deployment in contexts where alignment with human priorities is critical.