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

ICLR 2026 Conference Submission13745 Authors

Keywords: Ethical AI, Bayesian Experimental Design, System-level Evaluation

TL;DR: We propose SEED-SET, a scalable Bayesian framework for ethical testing that aligns system metrics with evolving human values using hierarchical models and LLM proxies.

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 both models.
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.

Primary Area: alignment, fairness, safety, privacy, and societal considerations

Submission Number: 13745