Securing the AI Revolution: NSFOCUS LLM Security Protection Solution
As Artificial Intelligence technology rapidly advances, Large Language Models (LLMs) are being widely adopted across countless domains. However, with this...
LLM
The Role of Humans in an AI-Powered World
As AI capabilities grow, we must delineate the roles that should remain exclusively human. The line seems to be...
Germany’s BSI issues guidelines to counter evasion attacks targeting LLMs
Germany’s BSI issues guidelines to counter evasion attacks targeting LLMs Pierluigi Paganini November 14, 2025 Germany’s BSI warns of rising...
AI chat privacy at risk: Microsoft details Whisper Leak side-channel attack
AI chat privacy at risk: Microsoft details Whisper Leak side-channel attack Pierluigi Paganini November 09, 2025 Microsoft uncovered Whisper Leak,...
OpenAI’s Aardvark is an AI Security Agent Combating Code Vulnerabilities
OpenAI on Thursday launched Aardvark, an artificial intelligence (AI) agent designed to autonomously detect and help fix security vulnerabilities...
AI Application Security with MCP Server | Contrast MCP Server | Contrast Security
When we introduced the Contrast Model-Context Protocol (MCP) Server a few months ago (read Supercharge your vulnerability remediation with...
Locking it down: A new technique to prevent LLM jailbreaks
Many organizations are increasingly deploying large language models (LLMs) such as OpenAI’s GPT series, Anthropic’s Claude, Meta’s LLaMA, and various...
Getting salty with LLMs: SophosAI unveils new defense against jailbreaking at CAMLIS 2025
Scientists from the SophosAI team will present their research at the upcoming Conference on Applied Machine Learning in Information Security...
Agentic AI’s OODA Loop Problem
The OODA loop—for observe, orient, decide, act—is a framework to understand decision-making in adversarial situations. We apply the same framework...
AI and the Future of American Politics
Two years ago, Americans anxious about the forthcoming 2024 presidential election were considering the malevolent force of an election influencer:...
Researchers expose MalTerminal, an LLM-enabled malware pioneer
Researchers expose MalTerminal, an LLM-enabled malware pioneer Pierluigi Paganini September 22, 2025 SentinelOne uncovered MalTerminal, the earliest known malware with...
LunaLock Ransomware threatens victims by feeding stolen data to AI models
LunaLock Ransomware threatens victims by feeding stolen data to AI models Pierluigi Paganini September 09, 2025 LunaLock, a new ransomware...
Encrypt AI, Protect Your IP: DataKrypto Tackles the LLM Security Crisis While Redefining What Encryption Should Be!
In the crowded, chaotic energy we’ve come to expect from the epic annual RSA Conference, some of the most meaningful...
“Unforeseen Discrepancy” in Large Language Models
“Unforeseen Discrepancy” in LLMs A fascinating study: “Unforeseen Discrepancy: Limited fine-tuning can generate broadly misaligned LLMs“: Summary: We introduce a...
New Study Reveals AI Disobeying Regulations
Recent Observations on AI Violating Standards Researchers experimented with Language Learning Models playing chess against superior adversaries. In instances where...
