Prompt injection through log files: what red-teaming our assistant found
Article URL: https://jl3-cloud.site/blog/prompt-injection-through-log-files/ Comments URL: https://news.ycombinator.com/item?id=50046522 Points: 2 # Comments: 0
Article URL: https://jl3-cloud.site/blog/prompt-injection-through-log-files/ Comments URL: https://news.ycombinator.com/item?id=50046522 Points: 2 # Comments: 0
Large language model (LLM) agents are moving from chat interfaces into infrastructure operations, where they read telemetry, call tools, generate and execute code, and change the state of production Kubernetes clusters. This collapses a boundary that conventional cloud-native security assumes: the boundary between data and control. Content that an agent merely reads (a log line, a ticket, a tool description) can redirect what it does. This paper argues that the model must…
Abstract Accurate cytological evaluation of body fluid specimens is essential for diagnosing inflammatory and malignant conditions, yet manual interpretation remains labor-intensive and subject to inter-observer variability. This study developed and deployed a smartphone-based deep learning framework for multi-class body fluid cell classification integrating object detection, ROI-based classification, and explainable artificial intelligence (XAI). Three hundred…
Smartwatches and other wearable IoT devices constantly stream physiological and behavioral data, making them vulnerable to zero-day attacks that traditional signature-based intrusion detection systems (IDS) are unable to identify and that current hybrid edge-cloud IDS leave momentarily unprotected while a cloud verdict is awaited. This approach outlines a fully reproducible three-tier detection pipeline: publisher, edge, and cloud. An edge-resident Isolation Forest scores…
This paper proposes Sensor-Initialized Gaussian Splatting (SIGS), a method that uses panoramic depth data acquired by the LiDAR sensor in a smartphone for three-dimensional (3D) scene reconstruction and VR space generation. Traditionally, 3D scene generation has required specialized knowledge and significant time, posing challenges for its application in VR. In particular, point clouds estimated by Structure from Motion (SfM), which are used for initializing 3D Gaussian…
Publicly exposed large language model (LLM) infrastructure creates a growing attack surface, yet real-world targeting remains poorly understood. We present Ollure, a low- and medium-interaction honeypot that emulates the Ollama API without a backend LLM. Spanning four deployments across cloud and university networks, Ollure operated for 84 days and recorded 290,887 interactions from 2,793 unique source IP addresses. Most of the activity consisted of automated discovery…
By combining digital sensing, network connectivity, data-driven analyses, cloud services, and automated controls, smart farming has been increasingly adopted in agricultural production. Although these technologies have improved the precision and efficiency of farm management, they also increase cybersecurity exposure as agricultural facilities are connected to external networks, platforms, and remote-control environments. This review seeks to clarify why cybersecurity should…
Article URL: https://cloud.google.com/blog/topics/threat-intelligence/from-prompting-to-autonomy-the-evolution-of-adversarial-ai Comments URL: https://news.ycombinator.com/item?id=49628034 Points: 13 # Comments: 1
Commercial LLM APIs advertise a specific foundation model, but the served backbone may be silently substituted, quantized, or wrapped, for example to save deployment costs. All existing audits decide backbone identity from the text-output channel, which is structurally fragile for agentic APIs because modern serving stacks (OpenAI, Anthropic, Gemini, Cloudflare Workers AI, LangGraph) discard text and expose only structured actions when the model calls a tool, and…
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…
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…
US employers and Haitian immigrants live in constant uncertainty as status of work permits remains in limbo For more than a month, Bobby Kuchinsky arrived at his Florida restaurant each morning worried about whether he still had enough staff or if this might be the last shift for some of his Haitian workers, who have spent more than a decade as cooks, busboys and oyster shuckers. On Tuesday, he suspended six of them after they lost authorization to work in the US.
Cloud telemetry arrives at a scale that, paradoxically, makes intrusion understanding harder rather than easier. Attackers operate through legitimate identity, federated session tokens, and cloud native APIs indistinguishable from routine administration, and analysts spend an incident reconstructing context the logs already contain. We present Cloud Decoy AI Agent, a framework pairing a high fidelity cloud decoy with an autonomous language model agent that compresses the…
Agentic systems now act with user secrets for APIs, messaging platforms, and cloud services. Today's agent runtimes typically implement authorization by exposure: enabling action means placing a reusable secret, or a reusable artifact derived from it, inside the runtime, so a transient prompt-injection or tool-side compromise becomes durable account compromise. Existing defenses cover adjacent pieces such as secret storage, scoped delegation, sender-constrained tokens, and…
LLM debugging agents that consume cloud logs and execute remediation commands are vulnerable to indirect prompt injection through log content. We present LogJack, a benchmark of 42 payloads across 5 cloud log categories, and evaluate 8 foundation models under 3 prompt conditions with 5 independent trials each (n = 160 per model per condition on 32 attack payloads). Under the active condition, verbatim command execution rates range from 0% (Claude Sonnet 4.6) to 86.2% (Llama…
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…
As cloud environments become increasingly complex, cybersecurity and forensic investigations must evolve to meet emerging threats. Large Language Models (LLMs) have shown promise in automating log analysis and reasoning tasks, yet they remain vulnerable to prompt injection attacks and lack forensic rigor. To address these dual challenges, we propose a unified, secure-by-design GenAI framework that integrates PromptShield and the Cloud Investigation Automation Framework…
Deploying large language models (LLMs) as autonomous browser agents exposes a significant attack surface in the form of Indirect Prompt Injection (IPI). Cloud-based defenses can provide strong semantic analysis, but they introduce latency and raise privacy concerns. We present the Cognitive Firewall, a three-stage split-compute architecture that distributes security checks across the client and the cloud. The system consists of a local visual Sentinel, a cloud-based Deep…
Phishing sites continue to increase in number and sophistication. Recent work uses large language models (LLMs) to analyze URLs, HTML, and rendered content and determine whether a website is a phishing site. However, these systems also introduce a new attack surface: prompt injection. Because attackers control many elements of a phishing site, they can embed malicious instructions that exploit perceptual asymmetry between LLMs and human users. Content that goes unnoticed by…
I just had my first success using a browser agent - in this case the Claude in Chrome extension - to solve an actual problem. A while ago I set things up so anything served from the https://static.simonwillison.net/static/cors-allow/ directory of my S3 bucket would have open Access-Control-Allow-Origin: * headers. This is useful for hosting files online that can be loaded into web applications hosted on other domains. Problem is I couldn't remember how I did it! I initially…