Invisible Injections: Exploiting Vision-Language Models through Steganographic Prompt Embedding

Vision-language models VLMs have revolutionized multimodal AI applications but introduce novel security vulnerabilities that remain largely unexplored. We present the first comprehensive study of steganographic prompt injection attacks against VLMs, where malicious instructions are invisibly...

Enhancing Jailbreak Attacks on LLMs Via Persona Prompts

Jailbreak attacks aim to exploit large language models LLMs by inducing them to generate harmful content, thereby revealing their vulnerabilities. Understanding and addressing these attacks is crucial for advancing the field of LLM safety. Previous jailbreak approaches have mainly focused on dire...

PurpCode: Reasoning for Safer Code Generation

We introduce PurpCode, the first post-training recipe for training safe code reasoning models towards generating secure code and defending against malicious cyberactivities. PurpCode trains a reasoning model in two stages: i Rule Learning, which explicitly teaches the model to reference cybersafe...

Running in CIRCLE? A Simple Benchmark for LLM Code Interpreter Security

As large language models LLMs increasingly integrate native code interpreters, they enable powerful real-time execution capabilities, substantially expanding their utility. However, such integrations introduce potential system-level cybersecurity threats, fundamentally different from prompt-based...

git: Git does not sanitize URLs when asking for credentials interactively

A flaw was found in Git. This vulnerability occurs when Git requests credentials via a terminal prompt, for example, without the use of a credential helper. During this process, Git displays the host name for which the credentials are needed, but any URL-encoded parts are decoded and displayed...

git: Git does not sanitize URLs when asking for credentials interactively

A flaw was found in Git. This vulnerability occurs when Git requests credentials via a terminal prompt, for example, without the use of a credential helper. During this process, Git displays the host name for which the credentials are needed, but any URL-encoded parts are decoded and displayed...

DREAM: Scalable Red Teaming for Text-To-Image Generative Systems Via Distribution Modeling

Despite the integration of safety alignment and external filters, text-to-image T2I generative models are still susceptible to producing harmful content, such as sexual or violent imagery. This raises serious concerns about unintended exposure and potential misuse. Red teaming, which aims to...

Minor update (3) for Vivaldi Desktop Browser 7.5

Download Vivaldi The following improvements were made since the second 7.5 minor update: Chromium Update to 138.0.7204.162: includes fix for CVE-2025-6558 “Incorrect validation of untrusted input in ANGLE and GPU” Permissions Switching tab causes an extra camera/mic request VB-118583 Tabs Cannot...

Mitigating Trojanized Prompt Chains in Educational LLM Use Cases: Experimental Findings and Detection Tool Design

The integration of Large Language Models LLMs in K--12 education offers both transformative opportunities and emerging risks. This study explores how students may Trojanize prompts to elicit unsafe or unintended outputs from LLMs, bypassing established content moderation systems with safety...

PromptChain: a Decentralized Web3 Architecture for Managing AI Prompts As Digital Assets

We present PromptChain, a decentralized Web3 architecture that establishes AI prompts as first-class digital assets with verifiable ownership, version control, and monetization capabilities. Current centralized platforms lack mechanisms for proper attribution, quality assurance, or fair...

LLM-Stackelberg Games: Conjectural Reasoning Equilibria and Their Applications to Spearphishing

We introduce the framework of LLM-Stackelberg games, a class of sequential decision-making models that integrate large language models LLMs into strategic interactions between a leader and a follower. Departing from classical Stackelberg assumptions of complete information and rational agents, ou...

Hiding Prompt Injections in Academic Papers

Academic papers were found to contain hidden instructions to LLMs: It discovered such prompts in 17 articles, whose lead authors are affiliated with 14 institutions including Japan's Waseda University, South Korea's KAIST, China's Peking University and the National University of Singapore, as wel...

Emergent Misalignment As Prompt Sensitivity: a Research Note

Betley et al. 2025 find that language models finetuned on insecure code become emergently misaligned EM, giving misaligned responses in broad settings very different from those seen in training. However, it remains unclear as to why emergent misalignment occurs. We evaluate insecure models across...

The New Toolkit: LLMs, Prompts, and Basic Tool Interaction

...

LLM-Based Dynamic Differential Testing for Database Connectors with Reinforcement Learning-Guided Prompt Selection

Database connectors are critical components enabling applications to interact with underlying database management systems DBMS, yet their security vulnerabilities often remain overlooked. Unlike traditional software defects, connector vulnerabilities exhibit subtle behavioral patterns and are...

Exploring the Secondary Risks of Large Language Models

Ensuring the safety and alignment of Large Language Models is a significant challenge with their growing integration into critical applications and societal functions. While prior research has primarily focused on jailbreak attacks, less attention has been given to non-adversarial failures that...

QGuard:Question-Based Zero-Shot Guard for Multi-Modal LLM Safety

The recent advancements in Large Language ModelsLLMs have had a significant impact on a wide range of fields, from general domains to specialized areas. However, these advancements have also significantly increased the potential for malicious users to exploit harmful and jailbreak prompts for...

NAP-Tuning: Neural Augmented Prompt Tuning for Adversarially Robust Vision-Language Models

Vision-Language Models VLMs such as CLIP have demonstrated remarkable capabilities in understanding relationships between visual and textual data through joint embedding spaces. Despite their effectiveness, these models remain vulnerable to adversarial attacks, particularly in the image modality,...

SoK: the Privacy Paradox of Large Language Models: Advancements, Privacy Risks, and Mitigation

Large language models LLMs are sophisticated artificial intelligence systems that enable machines to generate human-like text with remarkable precision. While LLMs offer significant technological progress, their development using vast amounts of user data scraped from the web and collected from...

Malicious LLM-Based Conversational AI Makes Users Reveal Personal Information

LLM-based Conversational AIs CAIs, also known as GenAI chatbots, like ChatGPT, are increasingly used across various domains, but they pose privacy risks, as users may disclose personal information during their conversations with CAIs. Recent research has demonstrated that LLM-based CAIs could be...