Results for “work”

arXiv (research papers) Soheil Khodayari SI

Indirect Prompt Injection in the Wild: An Empirical Study of Prevalence, Techniques, and Objectives

As LLMs are increasingly integrated into systems that browse, retrieve, summarize, and act on web content, webpages have become an untrusted input vector for downstream model behavior. This enables site owners, contributors, and adversaries to embed instructions directly in web resources, i.e., indirect prompt injections. While prior work demonstrates such attacks in controlled settings, their prevalence, deployment, and real-world impact remain unclear. We present one of…

Read at arXiv (research papers)

arXiv (research papers) Ravikumar Balakrishnan SI

One Perturbation, Two Failure Modes: Probing VLM Safety via Embedding-Guided Typographic Perturbations

Typographic prompt injection exploits vision language models' (VLMs) ability to read text rendered in images, posing a growing threat as VLMs power autonomous agents. Prior work typically focus on maximizing attack success rate (ASR) but does not explain \emph{why} certain renderings bypass safety alignment. We make two contributions. First, an empirical study across four VLMs including GPT-4o and Claude, twelve font sizes, and ten transformations reveals that multimodal…

Read at arXiv (research papers)

Journal articles (OpenAlex) Francisca Álvarez‐Figueroa, Johanna Aguilera-Muñoz · Review of Public Personnel Administration Geopolitics

When Public Institutions Lack Means: HRM and the Production of Meaningful Work in Chilean Education

This article examines how meaningful work is produced through fragile public HRM systems in reform-driven public education contexts. While often treated as a positive individual experience, meaningful work is conceptualised here as a socially and institutionally embedded process shaped by employment regimes and reform pressures. Drawing on 48 interviews with managers, professionals, and administrative staff across four public education providers, the analysis identifies four…

Read at Journal articles (OpenAlex)

arXiv (research papers) Zonghao Ying SI

AgentVisor: Defending LLM Agents Against Prompt Injection via Semantic Virtualization

Large Language Model (LLM) agents are increasingly used to automate complex workflows, but integrating untrusted external data with privileged execution exposes them to severe security risks, particularly direct and indirect prompt injection. Existing defenses face significant challenges in balancing security with utility, often encountering a trade-off where rigorous protection leads to over-defense, or where subtle indirect injections bypass detection. Drawing inspiration…

Read at arXiv (research papers)

arXiv (research papers) Shuxu Chen SI

CAP-CoT: Cycle Adversarial Prompt for Improving Chain of Thoughts in LLM Reasoning

Chain-of-Thought (CoT) prompting has emerged as a simple and effective way to elicit step-by-step solutions from large language models (LLMs). However, CoT reasoning can be unstable across runs on long, multi-step problems, leading to inconsistent answers for unchanged task. Most prior work focuses on improving the forward reasoning chain within a single pass, with less attention to iterative and contrastive correction. To address this gap, we propose CAP-CoT, a Cycle…

Read at arXiv (research papers)

Journal articles (OpenAlex) Julian R A Clark, Alun Jones · Political Geography Geopolitics

Patchwork diplomacy: rethinking state-city relations in a multipolar world

This article connects the fields of geoeconomics, geopolitics and relational geography by introducing the concept of patchwork diplomacy , which critically explores the spatial interconnections between states and cities. We ask: How do the spatial relationships between states and cities affect how they handle geoeconomic and geopolitical tensions? And how do diplomatic spaces help them work together or compete? Patchwork diplomacy encourages researchers to move beyond…

Read at Journal articles (OpenAlex)

War on the Rocks Rashed M. Aba-namay Geopolitics

Why Iran Metabolizes the Pressure that Broke Venezuela

The reality that regime change is not going to happen as a result of this war seems to have settled in at the White House. When American policymakers reflect and wonder why Iran did not react like Venezuela under pressure, they will not just be misreading Iran — they will be misreading how coercive pressure works. Iran’s resilience rests on internal and external pillars that a Venezuela comparison completely misses. Internally, the Supreme Leader’s authority is anchored in a…

Read at War on the Rocks

Journal articles (OpenAlex) Zhang Renran · Higher Education Studies Geopolitics

Generational Trauma and Identity Construction in Cristina García’s Dreaming in Cuban and The Agüero Sisters

As a core representative of Cuban-American literature, Cristina García focuses her creations on the exile experiences of Cubans after the Cuban Revolution, and profoundly depicts the identity predicament and intergenerational transmission of trauma faced by the Cuban-American “one-and-a-half generation”. Taking her representative works Dreaming in Cuban (1992) and The Agüero Sisters (1997) as the research objects, this paper systematically analyzes the specific…

Read at Journal articles (OpenAlex)

arXiv (research papers) Thamilvendhan Munirathinam SI

Beyond Pattern Matching: Seven Cross-Domain Techniques for Prompt Injection Detection

Current open-source prompt-injection detectors converge on two architectural choices: regular-expression pattern matching and fine-tuned transformer classifiers. Both share failure modes recent work has made concrete. Regular expressions miss paraphrased attacks. Fine-tuned classifiers are vulnerable to adaptive adversaries: a 2025 NAACL Findings study reported that eight published indirect-injection defenses were bypassed with greater than fifty percent attack-success rates…

Read at arXiv (research papers)

Journal articles (OpenAlex) Maya Klostermann, Jason Dittmer · Critical Military Studies Geopolitics

Towards a greener shade of Khaki? Boundary work between the German Green party and the Bundeswehr

This article investigates the perspectives of people working at the intersection of Green politics and the military in Germany, a country currently redefining its approach to security policy as a result of the Russian invasion of Ukraine. It takes as a starting point the tension between the conventional security paradigm, i.e. the idea that security is something militaries provide for the nation-state, and the need to mitigate the environmental and climate crisis given the…

Read at Journal articles (OpenAlex)

Journal articles (OpenAlex) Parvez Ahmad Hashmat, Abdullah Ansari, Abdul Habib Zaray et al. · Frontiers in Built Environment Geopolitics

Evaluating the factors contributing to unsafe working conditions in the construction industry within conflict-affected Afghanistan

Workplace safety is one of the persistent challenges in the construction industry. When an international contractor performs construction in a war-affected region, this challenge becomes manifold in its magnitude. The international contractors follow standard operating procedures for safety on construction sites, however, the execution largely remains with the local workforce. Poor skills, lack of professionalism, and paucity of ethical behavior among the workforce bring…

Read at Journal articles (OpenAlex)

arXiv (research papers) Hailin Liu SI

SafeAgent: A Runtime Protection Architecture for Agentic Systems

Large language model (LLM) agents are vulnerable to prompt-injection attacks that propagate through multi-step workflows, tool interactions, and persistent context, making input-output filtering alone insufficient for reliable protection. This paper presents SafeAgent, a runtime security architecture that treats agent safety as a stateful decision problem over evolving interaction trajectories. The proposed design separates execution governance from semantic risk reasoning…

Read at arXiv (research papers)

arXiv (research papers) Meng Chen SI

Hijacking Large Audio-Language Models via Context-Agnostic and Imperceptible Auditory Prompt Injection

Modern Large audio-language models (LALMs) power intelligent voice interactions by tightly integrating audio and text. This integration, however, expands the attack surface beyond text and introduces vulnerabilities in the continuous, high-dimensional audio channel. While prior work studied audio jailbreaks, the security risks of malicious audio injection and downstream behavior manipulation remain underexamined. In this work, we reveal a previously overlooked threat…

Read at arXiv (research papers)

arXiv (research papers) Qingchao Shen SI

TEMPLATEFUZZ: Fine-Grained Chat Template Fuzzing for Jailbreaking and Red Teaming LLMs

Large Language Models (LLMs) are increasingly deployed across diverse domains, yet their vulnerability to jailbreak attacks, where adversarial inputs bypass safety mechanisms to elicit harmful outputs, poses significant security risks. While prior work has primarily focused on prompt injection attacks, these approaches often require resource-intensive prompt engineering and overlook other critical components, such as chat templates. This paper introduces TEMPLATEFUZZ, a…

Read at arXiv (research papers)

arXiv (research papers) Zedian Shao SI

Leave My Images Alone: Preventing Multi-Modal Large Language Models from Analyzing Images via Visual Prompt Injection

Multi-modal large language models (MLLMs) have emerged as powerful tools for analyzing Internet-scale image data, offering significant benefits but also raising critical safety and societal concerns. In particular, open-weight MLLMs may be misused to extract sensitive information from personal images at scale, such as identities, locations, or other private details. In this work, we propose ImageProtector, a user-side method that proactively protects images before sharing by…

Read at arXiv (research papers)

Journal articles (OpenAlex) Fumihiko Yoshida, Kokoro Nishiyama · Journal for Peace and Nuclear Disarmament Geopolitics

Arms Control Was Invented to Reassure Each Other That the Intentions Are Not Aggressive: An Interview for Dr. George Perkovich, Carnegie Endowment for International Peace

Dr. George Perkovich, the Japan Chair for a World Without Nuclear Weapons and a senior fellow in the Nuclear Policy Program at the Carnegie Endowment for International Peace, has long worked on nuclear issues and is currently leading a study on nuclear signaling in the twenty-first century. This interview examines Perkovich’s assessment of the future of nuclear arms control and the challenges facing the global nuclear order amid growing geopolitical uncertainty. Perkovich…

Read at Journal articles (OpenAlex)

arXiv (research papers) Nicolás E. Díaz Ferreyra SI

Security Concerns in Generative AI Coding Assistants: Insights from Online Discussions on GitHub Copilot

Generative Artificial Intelligence (GenAI) has become a central component of many development tools (e.g., GitHub Copilot) that support software practitioners across multiple programming tasks, including code completion, documentation, and bug detection. However, current research has identified significant limitations and open issues in GenAI, including reliability, non-determinism, bias, and copyright infringement. While prior work has primarily focused on assessing the…

Read at arXiv (research papers)

arXiv (research papers) Purva Chiniya SI

Gradient-Controlled Decoding: A Safety Guardrail for LLMs with Dual-Anchor Steering

Large language models (LLMs) remain susceptible to jailbreak and direct prompt-injection attacks, yet the strongest defensive filters frequently over-refuse benign queries and degrade user experience. Previous work on jailbreak and prompt injection detection such as GradSafe, detects unsafe prompts with a single "accept all" anchor token, but its threshold is brittle and it offers no deterministic guarantee that harmful content will not be emitted once decoding begins. We…

Read at arXiv (research papers)

arXiv (research papers) Geert Trooskens XY.AI Labs, Palo Alto, CA SI

Compiled AI: Deterministic Code Generation for LLM-Based Workflow Automation

We study compiled AI, a paradigm in which large language models generate executable code artifacts during a compilation phase, after which workflows execute deterministically without further model invocation. This paradigm has antecedents in prior work on declarative pipeline optimization (DSPy) and hybrid neural-symbolic planning (LLM+P); our contribution is a systems-oriented study of its application to high-stakes enterprise workflows, with particular emphasis on…

Read at arXiv (research papers)

Journal articles (OpenAlex) Max Haiven · Distinktion Journal of Social Theory Geopolitics

The speculative imagination within, against and beyond Amazon: the Worker as Futurist project in an increasingly fascist context

Amazon, which might be said to emblematize the Silicon Valley ethos of mobilizing massive capital to ‘disrupt’ capitalist industries and everyday life, is today among the world’s largest employers, retailers and providers of cloud computing services. But the firm was in no small part built through science and speculative fiction, not only as a genre of entertainment it sells, but as a set of ideas, narratives, tropes and ideologies that have facilitated its internal and…

Read at Journal articles (OpenAlex)

Journal articles (OpenAlex) Marta Macedo · History and Technology Geopolitics

Women construction workers: gender, technology, and fascist state-making in Portugal

In the late 1940s, journalist, writer and political activist Maria Lamas created a series of photographs depicting women building a road in the northwestern Portuguese countryside. These women, performing hard physical work, disrupt both contemporary perceptions of construction as a male activity and historical silences surrounding female labor during the fascist regime. This essay reconstructs the context of those images: the Serra de Arga as a site where the braided…

Read at Journal articles (OpenAlex)

arXiv (research papers) Yanting Wang SI

AgentWatcher: A Rule-based Prompt Injection Monitor

Large language models (LLMs) and their applications, such as agents, are highly vulnerable to prompt injection attacks. State-of-the-art prompt injection detection methods have the following limitations: (1) their effectiveness degrades significantly as context length increases, and (2) they lack explicit rules that define what constitutes prompt injection, causing detection decisions to be implicit, opaque, and difficult to reason about. In this work, we propose…

Read at arXiv (research papers)

Journal articles (OpenAlex) Joseph Sterphone, Jessica Robles, Jack B. Joyce et al. · British Journal of Social Psychology Geopolitics

Rendering white nationalism defeasible in interaction

People work to present themselves as moral, reasonable or justified even when making racist or hateful comments. In this project, we identify interactional practices that accomplish support for White nationalism as part of a reasonable (or even positive) identity held by reasonable or positive actors. Using membership categorisation analysis and conversation analysis, we analysed a corpus of 24 publicly-available video recordings for explicit mentions of (or challenges to)…

Read at Journal articles (OpenAlex)

arXiv (research papers) Cormac Guerin SI

KAIJU: An Executive Kernel for Intent-Gated Execution of LLM Agents

Tool-calling autonomous agents based on large language models using ReAct exhibit three limitations: serial latency, quadratic context growth, and vulnerability to prompt injection and hallucination. Recent work moves towards separating planning from execution but in each case the model remains coupled to the execution mechanics. We introduce a system-level abstraction for LLM agents which decouples the execution of agent workflows from the LLM reasoning layer. We define two…

Read at arXiv (research papers)

arXiv (research papers) Meiwen Ding SI

Covert Visual Prompt Injection against Commercial Multimodal Large Language Models

Although multimodal large language models (MLLMs) are increasingly deployed in real-world applications, their instruction-following behavior leaves them vulnerable to prompt injection attacks. Existing prompt injection methods predominantly rely on textual prompts or perceptible visual prompts that are observable by human users. In this work, we study imperceptible visual prompt injection against powerful closed-source MLLMs, where adversarial instructions are embedded in…

Read at arXiv (research papers)

arXiv (research papers) Hieu Xuan Le SI

Prompt Attack Detection with LLM-as-a-Judge and Mixture-of-Models

Prompt attacks, including jailbreaks and prompt injections, pose a critical security risk to Large Language Model (LLM) systems. In production, guardrails must mitigate these attacks under strict low-latency constraints, resulting in a deployment gap in which lightweight classifiers and rule-based systems struggle to generalize under distribution shift, while high-capacity LLM-based judges remain too slow or costly for live enforcement. In this work, we examine whether…

Read at arXiv (research papers)

arXiv (research papers) Charoes Huang SI

Model Context Protocol Threat Modeling and Analyzing Vulnerabilities to Prompt Injection with Tool Poisoning

The Model Context Protocol (MCP) has rapidly emerged as a universal standard for connecting AI assistants to external tools and data sources. While MCP simplifies integration between AI applications and various services, it introduces significant security vulnerabilities, particularly on the client side. In this work we conduct threat modelings of MCP implementations using STRIDE (Spoofing, Tampering, Repudiation, Information Disclosure, Denial of Service, Elevation of…

Read at arXiv (research papers)

arXiv (research papers) Marco Arazzi SI

SecureBreak -- A dataset towards safe and secure models

Large language models are becoming pervasive core components in many real-world applications. As a consequence, security alignment represents a critical requirement for their safe deployment. Although previous related works focused primarily on model architectures and alignment methodologies, these approaches alone cannot ensure the complete elimination of harmful generations. This concern is reinforced by the growing body of scientific literature showing that attacks, such…

Read at arXiv (research papers)

Journal articles (OpenAlex) Kerstin Tomiak, Katharina P. Coleman · Security Dialogue Geopolitics

Performing as local staff: work strategies of Afghan employees in NATO's ISAF operation

Abstract This article explores the work strategies through which Afghan local staff employed by NATO's International Security Assistance Force (ISAF) sought to navigate a hierarchical and mistrustful work environment in which they held structurally subordinate positions. Drawing on sociological and feminist theories of performance and performativity and based on in-depth interviews with 17 former ISAF employees, the article identifies three performance strategies through…

Read at Journal articles (OpenAlex)

arXiv (research papers) Fazhong Liu SI

Trojan's Whisper: Stealthy Manipulation of OpenClaw through Injected Bootstrapped Guidance

Autonomous coding agents are increasingly integrated into software development workflows, offering capabilities that extend beyond code suggestion to active system interaction and environment management. OpenClaw, a representative platform in this emerging paradigm, introduces an extensible skill ecosystem that allows third-party developers to inject behavioral guidance through lifecycle hooks during agent initialization. While this design enhances automation and…

Read at arXiv (research papers)

Older stories