SEED-SET: Scalable Evolving Experimental Design for System-level Ethical Testing

2026-03-02

AI Summary

This study addresses the lack of scalable, systematic methods for evaluating ethical alignment in high-risk autonomous systems—such as drones—within human-centered environments. The authors propose a Bayesian experimental design framework that, for the first time, unifies stakeholders’ subjective values with objective domain-specific metrics through a hierarchical Gaussian process and a novel acquisition strategy. This approach enables an interpretable trade-off between exploration and exploitation in high-dimensional test spaces by integrating active learning with multi-objective optimization. Empirical results demonstrate substantial improvements in both testing efficiency and coverage: across two autonomous agent ethics evaluation tasks, the method generates twice as many valid test cases as baseline approaches and achieves a 1.25× increase in high-dimensional space coverage.

Technology Category

• Philosophy and Ethics of AI: Safety, Robustness & Trustworthiness
• Humans and AI: Learning Human Values and Preferences
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• Economics, Online Markets and Human Computation: Fairness and ethical considerations in crowd work and in human-in-the-loop AI systems
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

As autonomous systems such as drones, become increasingly deployed in high-stakes, human-centric domains, it is critical to evaluate the ethical alignment since failure to do so imposes imminent danger to human lives, and long term bias in decision-making. Automated ethical benchmarking of these systems is understudied due to the lack of ubiquitous, well-defined metrics for evaluation, and stakeholder-specific subjectivity, which cannot be modeled analytically. To address these challenges, we propose SEED-SET, a Bayesian experimental design framework that incorporates domain-specific objective evaluations, and subjective value judgments from stakeholders. SEED-SET models both evaluation types separately with hierarchical Gaussian Processes, and uses a novel acquisition strategy to propose interesting test candidates based on learnt qualitative preferences and objectives that align with the stakeholder preferences. We validate our approach for ethical benchmarking of autonomous agents on two applications and find our method to perform the best. Our method provides an interpretable and efficient trade-off between exploration and exploitation, by generating up to $2\times$ optimal test candidates compared to baselines, with $1.25\times$ improvement in coverage of high dimensional search spaces.

Problem

• ethical alignment
• autonomous systems
• ethical benchmarking
• stakeholder preferences
• evaluation metrics

Innovation

• Bayesian experimental design
• ethical benchmarking
• hierarchical Gaussian Processes
• autonomous systems
• preference-aware testing