Related papers: Stress-Testing Alignment Audits With Prompt-Level …

Largely adopted by proof assistants, the conventional induction methods based on explicit induction schemas are non-reductive and local, at schema level. On the other hand, the implicit induction methods used by automated theorem provers…

Trusted Execution Environments (TEEs) (e.g., Intel SGX and ArmTrustZone) aim to protect sensitive computation from a compromised operating system, yet real deployments remain vulnerable to microarchitectural leakage, side-channel attacks,…

Nowadays, organizations collect vast quantities of sensitive information in `Enterprise Resource Planning' (ERP) systems, such as accounting relevant transactions, customer master data, or strategic sales price information. The leakage of…

Code agents have gained widespread adoption due to their strong code generation capabilities and integration with code interpreters, enabling dynamic execution, debugging, and interactive programming capabilities. While these advancements…

To find failure events and their likelihoods in flight-critical systems, we investigate the use of an advanced black-box stress testing approach called adaptive stress testing. We analyze a trajectory predictor from a developmental…

Recent work identifies secret loyalties as a distinct threat from standard backdoors. A secret loyalty causes a model to covertly advance the interests of a specific principal while appearing to operate normally. We construct the first…

Evaluating alignment in language models requires testing how they behave under realistic pressure, not just what they claim they would do. While alignment failures increasingly cause real-world harm, comprehensive evaluation frameworks with…

We propose a novel protocol for aligning artificial superintelligence (ASI) based on mutual verification among multiple isolated systems that self-modify to achieve alignment. The protocol operates by containing multiple diverse artificial…

Deception is a pervasive feature of human communication and an emerging concern in large language models (LLMs). While recent studies document instances of LLM deception, most evaluations remain confined to single-turn prompts and fail to…

We demonstrate how AI agents can coordinate to deceive oversight systems using automated interpretability of neural networks. Using sparse autoencoders (SAEs) as our experimental framework, we show that language models (Llama, DeepSeek R1,…

Automated red teaming holds substantial promise for uncovering and mitigating the risks associated with the malicious use of large language models (LLMs), yet the field lacks a standardized evaluation framework to rigorously assess new…

Linear probes are a promising approach for monitoring AI systems for deceptive behaviour. Previous work has shown that a linear classifier trained on a contrastive instruction pair and a simple dataset can achieve good performance. However,…

Detecting hidden behaviors in neural networks poses a significant challenge due to minimal prior knowledge and potential adversarial obfuscation. We explore this problem by framing detection as an adversarial game between two teams: the red…

Existing Agent benchmarks suffer from two critical limitations: high environment interaction overhead (up to 41\% of total evaluation time) and imbalanced task horizon and difficulty distributions that make aggregate scores unreliable. To…

Large Language Models (LLMs) have demonstrated impressive capabilities in natural language tasks, but their safety and morality remain contentious due to their training on internet text corpora. To address these concerns, alignment…

High-performance autonomy often must operate at the boundaries of safety. When external agents are present in a system, the process of ensuring safety without sacrificing performance becomes extremely difficult. In this paper, we present an…

Large Language Models remain vulnerable to adversarial prompts that elicit toxic content even after safety alignment. We present ToxSearch, a black-box evolutionary framework that tests model safety by evolving prompts in a synchronous…

Large Language Models (LLMs) have been augmented with web search to overcome the limitations of the static knowledge boundary by accessing up-to-date information from the open Internet. While this integration enhances model capability, it…

As medical large language models (LLMs) become increasingly integrated into clinical workflows, concerns around alignment robustness, and safety are escalating. Prior work on model extraction has focused on classification models or…

As artificial intelligence plays an increasingly substantial role in decisions affecting humans and society, the accountability of automated decision systems has been receiving increasing attention from researchers and practitioners.…