Quality-Diversity Evolution for Discovering Diverse Vulnerabilities in LLM Safety

Current approaches to LLM adversarial testing suffer from coverage gaps: manual red-teaming does not scale, LLM-as-attacker methods exhibit mode collapse, and gradient-based approaches produce uninterpretable gibberish. We introduce a quality-diversity evolutionary framework that operates at the...

Context-Aware Spear Phishing: Generative AI-Enabled Attacks against Individuals Via Public Social Media Data

We demonstrate how publicly available social-media data and generative AI GenAI can be misused to automate and scale highly personalized, context-aware spear-phishing campaigns. With minimal attacker effort, a small amount of public activity per target is sufficient for GenAI models to extract...

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...

The Role of Learning in Attacking Intrusion Detection Systems

Recent work on network attacks have demonstrated that ML-based network intrusion detection systems NIDS can be evaded with adversarial perturbations. However, these attacks rely on complex optimizations that have large computational overheads, making them impractical in many real-world settings. ...

Exposing Vulnerabilities in Counterfeit Prevention Systems Utilizing Physically Unclonable Surface Features

Counterfeit products pose significant risks to public health and safety through infiltrating untrusted supply chains. Among numerous anti-counterfeiting techniques, leveraging inherent, unclonable microscopic irregularities of paper surfaces is an accurate and cost-effective solution. Prior work ...

RunawayEvil: Jailbreaking the Image-To-Video Generative Models

Image-to-Video I2V generation synthesizes dynamic visual content from image and text inputs, providing significant creative control. However, the security of such multimodal systems, particularly their vulnerability to jailbreak attacks, remains critically underexplored. To bridge this gap, we...

RedCoder: Automated Multi-Turn Red Teaming for Code LLMs

Large Language Models LLMs for code generation i.e., Code LLMs have demonstrated impressive capabilities in AI-assisted software development and testing. However, recent studies have shown that these models are prone to generating vulnerable or even malicious code under adversarial settings...

Exploiting Efficiency Vulnerabilities in Dynamic Deep Learning Systems

The growing deployment of deep learning models in real-world environments has intensified the need for efficient inference under strict latency and resource constraints. To meet these demands, dynamic deep learning systems DDLSs have emerged, offering input-adaptive computation to optimize runtim...

Scattered Spider: Understanding Help Desk Scams and How to Defend Your Organization

In the wake of high-profile attacks on UK retailers Marks & Spencer and Co-op, Scattered Spider has been all over the media, with coverage spilling over into the mainstream news due to the severity of the disruption caused — currently looking like hundreds of millions in lost profits for M&S alon...

A Taxonomy of Attacks and Defenses in Split Learning

Split Learning SL has emerged as a promising paradigm for distributed deep learning, allowing resource-constrained clients to offload portions of their model computation to servers while maintaining collaborative learning. However, recent research has demonstrated that SL remains vulnerable to a...

WalletProbe: a Testing Framework for Browser-Based Cryptocurrency Wallet Extensions

Serving as the first touch point for users to the cryptocurrency world, cryptocurrency wallets allow users to manage, receive, and transmit digital assets on blockchain networks and interact with emerging decentralized finance DeFi applications. Unfortunately, cryptocurrency wallets have always...

Earth Preta Evolves its Attacks with New Malware and Strategies

In this blog entry, we discuss our analysis of Earth Preta’s enhancements in their attacks by introducing new tools, malware variants and strategies to their worm-based attacks and their time-sensitive spear-phishing campaign...

Microsoft Releases PyRIT - A Red Teaming Tool for Generative AI

Microsoft has released an open access automation framework called PyRIT short for Python Risk Identification Tool to proactively identify risks in generative artificial intelligence AI systems. The red teaming tool is designed to "enable every organization across the globe to innovate responsibly...

A Quick Guide for Small Cybersecurity Teams Looking to Invest in Cyber Insurance

In the world of insurance providers and policies, cyber insurance is a fairly new field. And many security teams are trying to wrap their heads around it. What is it and do they need it? And with what time will they spend researching how to integrate cyber insurance into their strategy? For small...

Combining AI and Playbooks to Predict Cyberattacks

When organizations invest in AI, they are not only able to automate menial tasks like patching, but they can also create an automated system that looks for and discovers attacks, not only after the fact, but even before they occur. This predictive capability becomes increasingly necessary as...

Power Grid Honeypot Puts Face on Attacks

TENERIFE, Spain –The rhetoric around hacking the power grid would have you believe it’s a relatively mundane practice. Policymakers, intelligence agencies and vendors, for example, spread the word gleefully, leaning on scenarios such as state-sponsored hackers shutting off the lights in the dead ...

Businesses Remain Scared of Spear-Phishing as Attackers Study Behavior

Scared is a strong word, but the reality, according to a Websense analysis by Patrik Runald, is that spear-phishers, like the ones that compromised a White House network last week, are implementing new evasion tactics, fundamentally changing their attack strategies, and revolutionizing the target...

Phoenix Exploit's Kit 2.8 mini version

Phoenix Exploit's Kit 2.8 mini version Back in April of this year, we reported the leak of Phoenix Exploit Kit 2.5. The version currently in circulation is 2.8, and despite its lower activity for the last half of this year, it remains one of the preferred exploit packs used by cyber-criminals. In...