Related papers: Securing Computer-Use Agents: A Unified Architectu…
Mobile agents research is clearly aiming towards imposing agent based development as the next generation of tools for writing software. This paper comes with its own contribution to this global goal by introducing a novel unifying framework…
Generative AI systems achieve impressive performance on standard benchmarks yet fail to deliver real-world utility, a disconnect we identify across 28 deployment cases spanning education, healthcare, software engineering, and law. We argue…
The arrival of large language models (LLMs) capable of multi-step reasoning, tool use, and long-horizon planning has produced a qualitative shift in software engineering. Where earlier code-completion tools such as GitHub Copilot operated…
Claude Code is an agentic coding tool that can run shell commands, edit files, and call external services on behalf of the user. This study describes its comprehensive architecture by analyzing the publicly available TypeScript source code…
To enable human oversight, agentic AI systems often provide a trace of reasoning and action steps. Designing traces to have an informative, but not overwhelming, level of detail remains a critical challenge. In three user studies on a…
In a world, where complexity increases on a daily basis the Function-as-a-Service (FaaS) cloud model seams to take countermeasures. In comparison to other cloud models, the fast evolving FaaS increasingly abstracts the underlying…
Assurance Cases (ACs) are an established approach in safety engineering to argue quality claims in a structured way. In the context of quality assurance for Machine Learning (ML)-based software components, ACs are also being discussed and…
Healthcare organizations are beginning to embed agentic AI into routine workflows, including clinical documentation support and early-warning monitoring. As these capabilities diffuse across departments and vendors, health systems face…
We report an exploratory red-teaming study of autonomous language-model-powered agents deployed in a live laboratory environment with persistent memory, email accounts, Discord access, file systems, and shell execution. Over a two-week…
Learned classifiers deployed in agentic pipelines face a fundamental reliability problem: predictions are probabilistic inferences, not verified conclusions, and acting on them without grounding in observable evidence leads to compounding…
Autonomous browsing agents powered by large language models (LLMs) are increasingly used to automate web-based tasks. However, their reliance on dynamic content, tool execution, and user-provided data exposes them to a broad attack surface.…
Traditional AI reasoning techniques have been used successfully in many domains, including logistics, scheduling and game playing. This paper is part of a project aimed at investigating how such techniques can be extended to coordinate…
The rise of OpenClaw in early 2026 marks the moment when millions of users began deploying personal AI agents into their daily lives, delegating tasks ranging from travel planning to multi-step research. This scale of adoption signals that…
Autonomous AI agents powered by large language models are being deployed in production with capabilities including shell execution, file system access, database queries, and multi-party communication. Recent red teaming research…
Agentic language-model systems increasingly rely on mutable execution contexts, including files, memory, tools, skills, and auxiliary artifacts, creating security risks beyond explicit user prompts. This paper presents DeepTrap, an…
As autonomous agentic AI systems see increasing adoption across organisations, persistent challenges in alignment, governance, and risk management threaten to impede deployment at scale. We present AURA (Agent aUtonomy Risk Assessment), a…
Repurposing large vision-language models (LVLMs) as computer use agents (CUAs) has led to substantial breakthroughs, primarily driven by human-labeled data. However, these models often struggle with novel and specialized software,…
Multi-agent systems built on large language models (LLMs) require many coordination choices that are difficult to fix a priori: which skill protocol to invoke, which agent role should perform a subtask, which model to bind to each role, how…
As LLM-based computer-use agents (CUAs) begin to autonomously interact with real-world interfaces, understanding their vulnerability to manipulative interface designs becomes increasingly critical. We introduce SusBench, an online benchmark…
LLM-powered Multi-Agent Systems (LLM-MAS) unlock new potentials in distributed reasoning, collaboration, and task generalization but also introduce additional risks due to unguaranteed agreement, cascading uncertainty, and adversarial…