Results for “security”
Delegation Without Trust: An Empirical Gap Analysis of Identity, Authorization, and Runtime Governance in Multi-Agent LLM Systems
Autonomous LLM agents increasingly act on a user's behalf: they hold credentials, call tools and services, and spawn sub-agents that act further on their behalf. This turns a long-standing distributed-systems question -- who is authorized to do what, on whose authority -- into an urgent and largely unsolved problem, because the component driving each agent is a language model an adversary can hijack. We argue that agent security must be evaluated under an untrusted-model…
Implementation of a Secured Smart Home System Using Consortium Blockchain on Raspberry Pi as an Edge Gateway
Smart homes are among the fastest-growing applications of the Internet of Things (IoT), enabling the automation and remote control of household devices such as door locks, lighting systems, fans, and security cameras. Ensuring the security of these interconnected devices is critical, as numerous global incidents have shown that traditional security mechanisms often fail to protect IoT systems from unauthorized access and cyberattacks. This paper presents a novel 4-layer…
The military spatial complex: A framework of neoliberal urban power relations
This paper develops the concept of the military–spatial complex (MSC) to analyze how armed forces increasingly shape neoliberal urban development. While military urbanism has often been studied through the visible apparatus of surveillance and security, the MSC highlights subtler forms of militarization: the embedding of martial authority into civilian urban life through planning, land governance, and property development. Correspondingly, this paper argues that militaries…
Haiti’s security crisis forces displacement and relocation
Gangs continue to terrorise the population, but a combination of armed civilian groups, police, and armed forces are trying to keep them at bay.
LongPIBench: A Long-Context Benchmark for Prompt Injection
Prompt injection attacks pose a serious security risk to large language models in real-world applications. However, existing prompt injection benchmarks primarily focus on short-context inputs, leaving the attacks and defenses in long-context settings largely unexplored. This gap leads to a substantial overestimation of the effectiveness of current defenses. In this paper, we bridge the gap by introducing LongPIBench, a long-context benchmark for prompt injection covering 4…
Venezuela’s earthquakes leave children facing trauma, displacement and food insecurity
As families seek refuge, Venezuela’s youngest children grapple with fear, disrupted routines and the loss of basic nutritional security.
An analytical framework for modeling security and drug trafficking supply chains as a complex system
Maritime drug trafficking in the Colombian Caribbean exhibits structural properties of a Complex Adaptive System (CAS), characterized by decentralized, evolving, and non-linear interactions. This study establishes a systems-based framework to analyze the adaptive behavior of illicit maritime logistics and extract intelligence for risk-based interdiction. A ten-attribute instrument was designed to quantify CAS dimensions, applied to a sample of 223 experts from military…
Will Kast’s Security Agenda Succeed at Making Chile Safer?
A Latin America Advisor Q&A featuring experts’ views on Chile’s and José Antonio Kast’s security agenda, and the prospects of getting the proposed security package through Congress.
The Black Sea grain corridor as a crisis-driven food security governance mechanism: evidence from Samsun, Türkiye
The Russia–Ukraine War severely disrupted global grain supply chains by blocking Ukrainian ports and interrupting exports from a major grain-producing region. In response, the Black Sea Grain Corridor (BSGC) was established in 2022 to restore maritime grain shipments and mitigate the emerging global food crisis. While existing research has largely focused on the Corridor’s effects on global prices and trade flows, limited attention has been paid to its localized economic…
Cheap Drones Are A Costly Problem For Governments in Latin America
Cheap Drones Are A Costly Problem For Governments in Latin America Mexico and the United States recently signed a non-binding “declaration of intent” deepening security cooperation around counter-drone technology, as governments across Latin America race to adapt to how criminal organizations are transforming the inexpensive commercial technology into increasingly sophisticated tools for surveillance, territorial control, and violence.
Security-by-Design and Risk-Based Certification for AI-Enabled Smart Home
The integration of Artificial Intelligence (AI) into Internet of Things (IoT) ecosystems has enabled the development of advanced cyber–physical systems, including smart appliances, while introducing security, privacy, and AI governance risks that extend beyond the scope of traditional threat models. Existing approaches often address cybersecurity, AI risk management, and regulatory compliance in isolation, leaving manufacturers without a systematic method for translating…
The Virtue of Bias? China’s Stakeholder Mediation in the Iran and North Korea Nuclear Crises
Why does China mediate differently across similar security crises? Challenging mediation theory’s neutrality assumption, this article conceptualises ‘stakeholder mediation’ as engagement by third parties whose security, economic, or reputational interests are exposed to a conflict’s outcome. Such mediators hold leverage but risk entrapment and reputational costs. Using a most-similar design, process tracing, diplomat memoirs, declassified documents, and elite interviews, we…
Kurdish-led Syrian Democratic Forces announce its dissolution
The dissolution of the Syrian Democratic Forces (SDF), the main US-backed anti-Islamic State group in Syria, ends more than a decade of Kurdish-led autonomy in the country’s northeast, as the SDF’s military, security, and civilian institutions are integrated into the state.
Egypt: After 100 Days, Omar Salah Marei’s arbitrary detention must end
We, the undersigned organisations, renew our call for the immediate and unconditional release of Egyptian filmmaker and screenwriter Omar Salah Marei and for an end to his pretrial detention in Case No. 3385 of 2026 (Supreme State Security Prosecution). Our appeal comes as Omar marks 100 days in detention and has now had his pretrial
WebMCP-Phalanx: Enforcing and Characterizing Trust Boundaries for Browser-Integrated LLM Agents
The emerging W3C WebMCP proposal enables LLM agents to invoke tools exposed by web pages. In multi-party web environments, however, integrating agent execution into a browser security model centered on the Same-Origin Policy (SOP) leaves insufficient provenance and lifecycle guarantees for agent-accessible tools, creating three risks: subject-attribution spoofing, uncontrolled tool lifecycles, and semantic prompt injection. We propose WebMCP-Phalanx, a dual-layer agent…
Marry the market: Romancing finance and redeeming capitalism in popular financial feminisms
This article examines the romancing of finance in and through popular financial feminisms, a transnational cultural terrain in which women claiming financial insight and expertise seek to ‘empower’ other women by sharing accessible and actionable advice. Such actors frequently promote stock market investing, claiming that engagement with financial markets represents the surest and most reliable means for women to achieve long-term financial security, alongside the various…
Beyond the Mandate: A Systematic Security Analysis of the Agent Payments Protocol (AP2)
The Agent Payments Protocol (AP2), introduced by Google, enables large language model (LLM)-driven shopping agents to authorize and execute payments on behalf of users. Its signed Checkout and Payment Mandates protect the integrity of transaction data after signing. Agent interactions and external inputs that shape a transaction before authorization remain outside that protection, including Agent-to-Agent Protocol (A2A) messages and Model Context Protocol (MCP) tool calls…
TraceGrant: A Contract-Governed Security Framework for the Task-Effect Lifecycle of Networked LLM Agents
Networked large language model (LLM) agents retrieve information from email, cloud storage, calendars, transaction platforms, and Web services to complete multistep tasks that produce persistent external effects. The same content needed for legitimate execution may also contain indirect prompt injections that redirect tool use, alter sensitive arguments, or disrupt task completion. Existing defenses mainly constrain untrusted content or individual tool calls, leaving user…
On the Radar: US Eyes Ground War Against Crime
On the Radar: US Eyes Ground War Against Crime What's next for the US military campaign against Latin America’s criminal groups? What’s behind Chile and Peru's recent security crackdowns? And are explosives and drones becoming standard weapons for criminal groups in the region? This week's top three organized crime stories in On the Radar.
A Colombia-Venezuela agenda for recovery, security, and economic growth
The recent earthquakes in Venezuela and Colombia have created urgent humanitarian needs that will require sustained international support. As recovery advances, the US-Colombia Strategic Alignment Coalition outlines how both countries, with US support, can build a cooperation agenda.
The Defence of Western Civilization: Racialized Liberalism and the American Commitment to the North Atlantic Security Community, 1947-1949
The North Atlantic Alliance has been a crucial pillar of American foreign policy since the end of the Second World War. According to conventional accounts, the imperative to defend “free people” from the menace of communism was critical in legitimizing the birth of the North Atlantic security community. However, why then was the defence of “free people” limited to European empires and not extended to populations segregated at home or those colonized by those very “Western…
An ECC- and OT-based privacy-preserving authentication scheme for smart home systems
Abstract Although the integration of 5G and AIoT technologies has created new opportunities for smart home systems, security issues remain a major barrier to their sustainable development and widespread adoption. Traditional identity authentication methods in smart home systems suffer from high computational complexity, complicated key management, and poor scalability, making them unsuitable for smart home environments with stringent storage and energy constraints. In…
Death and alleged torture of man in detention sparks protests in Syria
Picture showing White Helmets volunteer unconscious and with bruises on his body went viral Syrians have protested over the death of a man who was arrested and allegedly beaten by security forces, sparking widespread anger at a practice thought to have ended with the fall of Bashar al-Assad. A crowd gathered in the city of Aleppo, demanding accountability for the security forces who are alleged to have carried out the torture, with participants holding signs and slogans such…
Security of Foundation-Model-Powered Embodied Agents: Attack Surfaces, Attacks, Defenses, and Evaluation
Foundation models are increasingly used for perception, reasoning, planning, and action generation in embodied agents, creating security risks that can propagate from digital inputs to physical behavior. Existing surveys often organize threats by mechanisms such as jailbreaks, prompt injection, backdoors, poisoning, or adversarial examples, but these categories do not consistently identify where an adversary first enters the embodied control loop. We present a…
Security Assessment of DeepSeek Harness with A.I.G: Evaluating Resistance to Indirect Prompt Injection
We assess indirect prompt injection in DeepSeek Harness (DSH), using AI-Infra-Guard (A.I.G) to construct tests, deliver controlled taint, execute DSH, collect traces, and judge outcomes. The study covers 14,560 controlled executions over 16 indirect-content channels, text and file carrier modes, 35 payload objectives, one unmodified baseline, and 12 attack methods. The experiment preserves DSH's agent loop, tool registry, model adapter, and session-event path; source tools…
Bounded Agents: Delegation Security for Multi-Agent AI Systems
LLM-based agents can act on behalf of a user to access cloud services, call tools, or invoke agents. At session start, the agent's permissions are set but remain static, and each request is evaluated independently, without considering prior actions. Within its permissions, an agent may act contrary to the delegated task, combine individually permitted actions into a prohibited outcome, or delegate authority to a sub-agent without limiting it. A prompt injection poses a risk…
The Chilean Invisible Reconstruction in Post-Fire Recovery: Interdisciplinary Perspectives and Soil Security Challenges Amidst an Extractive Economic Model
Current post-fire recovery strategies in Mediterranean Chile primarily emphasize discernible outcomes, such as residential reconstruction and expedited reforestation. In this context, neglecting soil degradation remains a critical geographical oversight. This perspective proposes the integration of a soil security framework as an “invisible reconstruction” to mitigate the unacknowledged deterioration of the five soil dimensions (capacity, condition, capital, connectivity…
Rethinking Agent Security as a Networking Problem
AI agents are rapidly becoming more capable and widely deployed, promising substantial gains in productivity and enabling new classes of applications. However, their growing autonomy also introduces significant privacy and security risks. Existing defenses are predominantly agent-centric, relying on the agent itself to detect threats and enforce privacy and security policies. This approach is fundamentally limited because it entrusts policy enforcement to AI agents whose…