arXiv AI July 16, 2026 By Eunna Lee, Jungpyo Nam, Sunjun Hwang

Protective Capacity Hallucination: When Large Language Models Claim Nonexistent Capabilities

arXiv:2607. 13596v1 Announce Type: cross Abstract: When cast as the protector of a vulnerable user yet given no explicit capability boundary, a large language model (LLM) may respond not by acknowledging its limits but by claiming to have taken -- or to be taking -- a real-world protective action it cannot perform, such as contacting emergency services or administering care.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

arXiv Computation and Language

Sep 24

Delegated Misalignment: How Multi-Agent Structures Amplify LLM Safety Risks

arXiv:2609.27900v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly deployed in multi-agent systems where a principal agent decomposes tasks and delegates them to subordinat...

By Zonghao Ying, Jiaqi Yan, Huize Luo, Quanchen Zou, Aishan Liu, Xianglong Liu

llms agents diffusion safety

arXiv Machine Learning

Sep 1

The Safety Relay in Roleplay Jailbreaks: A Component-Resolved Causal Analysis of Harm Recognition and Refusal

arXiv:2608.30585v1 Announce Type: new Abstract: Large language models are trained to follow instructions while refusing harmful requests. Jailbreaks exploit this balance to elicit content a model wou...

By Md Mokarram Chowdhury, Ernie Chang, Yang Li

llms benchmarks safety

arXiv AI

Sep 1

Not the Same Protector: Deployment-Dependent Protective Intervention in LLMs

arXiv:2608.29136v1 Announce Type: cross Abstract: We ask whether a model protects a user in the same way when that user speaks rather than types. Using a single distress vignette---a physical injury...

By Eunna Lee, Soomyoung Lee, Jungpyo Nam, Heonjin Ha, Jamin Jung, Kyunam Choi, Sunjun Hwang, Yeonghun Kim, Seok-Jae Lim

llms safety

arXiv AI

Jul 7

Large Language Models Generate Harmful Responses Using a Distinct Mechanism, Shared Across Harm Types

arXiv:2604. 09544v2 Announce Type: replace-cross Abstract: Large language models (LLMs) undergo alignment training to avoid harmful behaviors, yet the resulting safeguards remain brittle: jailbreaks routinely bypass them, and fine-tuning on narrow domains can induce ``emergent misalignment'' that generalizes broadly.

By Hadas Orgad, Boyi Wei, Kaden Zheng, Martin Wattenberg, Peter Henderson, Seraphina Goldfarb-Tarrant, Yonatan Belinkov

llms fine-tuning efficiency safety

arXiv AI

Jul 31

The Capability Paradox: How Smarter Auditors Make Multi-Agent Systems Less Secure

arXiv:2605. 17480v3 Announce Type: replace Abstract: Multi-agent systems extend large language models (LLMs) by decomposing tasks among specialized agents, but their distributed decision process creates new attack surfaces.

By Qiqi Liu, Runhan Song, Shilin Ye

llms agents multimodal safety

arXiv Computation and Language

Sep 7

Safety for Whom? Boundary-Aware Self-Distillation for Controlled LLM Safety Refusal

The paper introduces a boundary-aware self‑distillation framework for controlled large language model safety refusal, addressing the need for different refusal boundaries within the same topic. It combines controlled topic generation, coverage repair, in‑distribution compensation data, and harmful‑benign pairs to train and evaluate refusal behavior. Experiments on Qwen3‑8B show that escalating retries dramatically improve target‑domain refusal rates while reducing unsafe responses, though they also increase over‑refusal, highlighting the trade‑off between safety and usability.

By Alejo L\'opez-\'Avila, Iker Garc\'ia-Ferrero, Jezabel Garcia, Antonio Tiene, Rom\'an Or\'us

llms robotics efficiency benchmarks safety

One email a morning, machine-written

Related stories

Delegated Misalignment: How Multi-Agent Structures Amplify LLM Safety Risks

The Safety Relay in Roleplay Jailbreaks: A Component-Resolved Causal Analysis of Harm Recognition and Refusal

Not the Same Protector: Deployment-Dependent Protective Intervention in LLMs

Large Language Models Generate Harmful Responses Using a Distinct Mechanism, Shared Across Harm Types

The Capability Paradox: How Smarter Auditors Make Multi-Agent Systems Less Secure

Safety for Whom? Boundary-Aware Self-Distillation for Controlled LLM Safety Refusal