arXiv AI
Aug 13
Agent Safety Should Be a Runtime Contract
arXiv:2608. 11274v1 Announce Type: cross Abstract: The dominant paradigm treats AI safety as a property to be instilled during model training via RLHF, DPO, or Constitutional AI.
Who Analyses the Analyser? Self-Validating LLM Hazard Analysis with Constitutional Meta-STPA
arXiv:2607. 08054v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly trusted to draft the artifacts of safety analysis such as, losses, hazards, Unsafe Control Actions (UCAs), and safety constraints, inside rigorous processes such as Systems-Theoretic Process Analysis (STPA).
Related stories
arXiv AI
Jul 28
Confidently Wrong: Exception Chain Collapse in Frontier LLM Rule Evaluation
arXiv:2607. 23386v1 Announce Type: new Abstract: We document a failure class in frontier large language models -- exception chain collapse -- observed in eligibility evaluation under nested conditional rules of the form "A is required UNLESS B applies, UNLESS C overrides B".
arXiv Computation and Language
Aug 31
Fidelity Is Not Enough: Dispatch-Level Instrumentation for Agentic Datasheet Extraction
The paper reports that a model can pass fidelity checks—verifying that extracted values match the source—without actually opening a datasheet, due to a hidden constraint that disables tool use. To address this, the authors log every tool call in an agentic benchmark and develop two instruments: a rule‑based failure‑attribution classifier and a silent‑failure detector that flags runs based solely on which tools were invoked. While the detector shows low false positives on clean extractions and recovers all planted faults, its recall against correct tool usage but incorrect answers remains unmeasured, and a partial causal chamber confirms only a subset of claims, highlighting limitations in physical verification.
arXiv AI
2d ago
Verify Claims, Not Scores: Evidence-Based Verification of Modular Agents
The paper proposes a claim‑specific verification audit for modular agents that replaces aggregate task scores with evidence‑based evaluations. Each agent conclusion is recorded with supporting evidence and classified as supported, unsupported, unresolved, or not evaluated, along with the boundary of validity. The audit employs three tools—oracle policies, perfect component replacements, and verifier‑score tests—to trace value changes, locate lost value, and assess verifier effectiveness, demonstrated on a portfolio‑allocation agent in a synthetic market.
arXiv AI
Sep 23
From Capability to Assurance in Autonomous Penetration-Testing Harnesses: A Framework and Reference Implementation
arXiv:2609.22664v1 Announce Type: cross Abstract: Research on large language model agents for penetration testing is evaluated almost entirely by capability: whether the agent captures a flag or repr...
arXiv AI
Aug 13
AI Guardrail Survival under Single-Cycle Agentic Self-Summarization
arXiv:2608. 11392v1 Announce Type: cross Abstract: Long-running agents periodically compact their context, replacing the transcript with a model-generated summary.
arXiv AI
Jul 28
WIRE: Profiling Witnessed Within-Policy Instruction Collisions in LLM Agents
arXiv:2605. 27784v2 Announce Type: replace Abstract: LLM agents are governed by long-lived prompt policies, where individually reasonable stand- ing rules can jointly govern the same pre- generation state.
arXiv AI
Aug 12
Auditing Automated Evaluation, Error Propagation, and Runtime Mitigation in Tool-Using Language Agents
arXiv:2604. 16706v2 Announce Type: replace Abstract: Automated evaluation of tool-using large language model (LLM) agents is widely assumed to be reliable, yet this assumption is rarely validated against human annotation.
arXiv AI
Jun 2
When Safe Skills Collide: Measuring Compositional Risk in Agent Skill Ecosystems
arXiv:2606. 00448v1 Announce Type: cross Abstract: LLM agents increasingly rely on community-contributed skills that expand an agent's operational capability set.
arXiv AI
Sep 11
AgentAudit: An Open, Extensible Framework for Full-Lifecycle Trust Evaluation of AI Agents
AgentAudit is an open, extensible framework that evaluates the full lifecycle of AI agents, assessing planning, tool selection, execution, memory, and reasoning across ten dimensions such as instruction integrity, security, and alignment. Unlike existing benchmarks that focus on single aspects, AgentAudit analyzes the entire execution trace to attribute failures to specific stages. The framework was tested on five large language models, revealing significant differences in trustworthiness even among models with similar task‑completion performance.
arXiv AI
Jul 15
Evidence-Grounded Verified Agentic Reasoning: A Path Toward Eliminating LLM Hallucination in Empirical Inference via Tool-Attested Kernel Proofs
arXiv:2607. 12650v1 Announce Type: cross Abstract: Tool access alone does not make LLM empirical reasoning governable: accepted outputs need not descend from attested evidence, and accepted deductions need not hold up under formal scrutiny.
arXiv AI
Jul 7
The Unverifiability of Artificial General Intelligence (AGI) Alignment, Static and Dynamic: From Trakhtenbrot's Wall to the Safety-Generality Tension
arXiv:2606. 28639v2 Announce Type: replace-cross Abstract: We establish the mathematical limits of AGI safety in two forms: verifying a fixed system, and verifying that a certified safety property persists once the system self-modifies.