Related papers: SpecEval: Evaluating Model Adherence to Behavior S…
Safety benchmark scores provide incomplete evidence of deployment readiness: aligned language models often adhere to rigid rules even when a situational update flips which action is safe. We term this failure brittle safety. To diagnose it,…
As frontier AI systems advance toward transformative capabilities, we need a parallel transformation in how we measure and evaluate these systems to ensure safety and inform governance. While benchmarks have been the primary method for…
Dialogue models are able to generate coherent and fluent responses, but they can still be challenging to control and may produce non-engaging, unsafe results. This unpredictability diminishes user trust and can hinder the use of the models…
Language models deployed in high-stakes professional settings face conflicting demands from users, institutional authorities, and professional norms. How models act when these demands conflict reveals a principal hierarchy -- an implicit…
An auditor instructs an AI assistant: "open each file individually using the Read tool -- no scripts, no agents." The AI replies "Yes" -- then issues a single batched call summarizing all fifty files at once. We call this the Compliance…
We are motivated by the problem of autonomous vehicle performance validation. A key challenge is that an autonomous vehicle requires testing in every kind of driving scenario it could encounter, including rare events, to provide a strong…
As AI systems become increasingly embedded in organizational workflows and consumer applications, ethical principles such as fairness, transparency, and robustness have been widely endorsed in policy and industry guidelines. However, there…
Model documentation plays a crucial role in promoting transparency and responsible development of AI systems. With the rise of Generative AI (GenAI), open-source platforms have increasingly become hubs for hosting and distributing these…
Typically, machine learning models are trained and evaluated without making any distinction between users (e.g, using traditional hold-out and cross-validation). However, this produces inaccurate performance metrics estimates in multi-user…
As Large Language Models (LLMs) are deployed with increasing real-world responsibilities, it is important to be able to specify and constrain the behavior of these systems in a reliable manner. Model developers may wish to set explicit…
Autonomous agents are increasingly being proposed for use in healthcare, assistive care, education, and other applications governed by complex human-centric norms. To ensure compliance with these norms, the rules they induce need to be…
Most frontier AI developers publicly document their safety evaluations of new AI models in model reports, including testing for chemical and biological (ChemBio) misuse risks. This practice provides a window into the methodology of these…
Machine learning models with high accuracy on test data can still produce systematic failures, such as harmful biases and safety issues, when deployed in the real world. To detect and mitigate such failures, practitioners run behavioral…
For autonomous vehicles, safe navigation in complex environments depends on handling a broad range of diverse and rare driving scenarios. Simulation- and scenario-based testing have emerged as key approaches to development and validation of…
The specification of requirements and tests are crucial activities in automotive development projects. However, due to the increasing complexity of automotive systems, practitioners fail to specify requirements and tests for distributed and…
Business analysts and domain experts are often sketching the behaviors of a software system using high-level models that are technology- and platform-independent. The developers will refine and enrich these high-level models with technical…
Function calling is a core capability of large language models, essential for AI agents. Existing benchmarks such as the Berkeley Function Calling Leaderboard (BFCL), tau^2-Bench (arXiv:2506.07982), and ACEBench (arXiv:2501.12851) evaluate…
Developers play a central role in determining how machine learning systems are explained in practice, yet they are rarely trained to design explanations for non-technical audiences. Despite this, transparency and explainability requirements…
Foundation models are increasingly central to high-stakes AI systems, and governance frameworks now depend on evaluations to assess their risks and capabilities. Although general capability evaluations are widespread, social impact…
Advances in service personalization are driven by low-cost data collection and processing, in addition to the wide variety of third-party frameworks for authentication, storage, and marketing. New privacy regulations, such as the General…