~/category/adversarial-ai
Attacks on and with AI — prompt injection, model red teaming, AI-driven offense and defense.
34 articles
Written by Anthony Giandomenicohttps://www.linkedin.com/in/anthonygiandomenico/The Shift: Not New, But Moving FastAgentic AI is not a new concept. At this point, most people in the industry have heard about it, and many have been experimenting with it in different forms. The idea of systems that can
Click here to listen on SoundCloud: https://soundcloud.com/drchaos-podcast/project-glasswing-podcastIn this week's episode of the Dr. Chaos Podcast, hosts Anthony Giandomenico (Tony G) and Aamir Lakhani (Dr. Chaos) dive deep into the rapidly evolving landscape of AI in cybersecurity. We are official
1. Model and AI Pipeline SecurityThe first area is what I’d call model and AI pipeline security.This is where a lot of the early focus has been — making sure the model itself, the data feeding it, and the overall pipeline are secure before anything ever gets deployed.These solutions are looking for
Human friendly Until It Isn'tAt first, they sounded perfectly human.Two AI agents exchanging polite, helpful sentences about booking a hotel. Dates. Cities. Room types. Small talk in clean, predictable English. Nothing unusual. Nothing alarming.And then something changed.A subtle handshake. A quiet
There’s a question that’s been sitting in the back of my mind lately.If we’re moving into a world where autonomous systems are planning, adapting, and executing actions with increasing independence… what does that actually mean for threat intelligence teams?I’m not asking this from a hype perspectiv
In conversations with people who are actually building and red teaming agentic systems, one theme keeps coming up: the failures they’re seeing don’t really look like prompt problems. They look like architectural ones.That framing matters, because it changes where attention goes. A lot of early work
For years, security teams learned how to protect systems that respond. Agentic AI introduces systems that decide.That difference sounds small at first. It isn’t.Most of today’s AI security conversation still treats models like tools: prompt in, response out, controls around the edges. Agentic syst
Check out a related podcast at https://www.drchaos.com/post/podcast-ai-agents-and-the-new-insider-threatAs we move into 2026, the cybersecurity landscape is undergoing a phase shift where AI is transitioning from a modular tool into an autonomous, operational actor For technical practitioners, this
Check out our companion blog at: https://www.drchaos.com/post/the-agentic-shift-architecting-for-the-crisis-of-authenticity-and-industrialized-cybercrime-in-2026Join our latest podcast. We explore the transformative impact of AI on insider threat dynamics. Learn about how AI agents manipulate system
My annual pilgrimage to Las Vegas for Black Hat USA and DEF CON—what many of my friends affectionately call “Hacker Summer Camp”—took place from August 2–7, 2025, at the Mandalay Bay Convention Center, with DEF CON extending the festivities through August 10 at the Las Vegas Convention Center.Last y
The first step is to update your system. Assuming you have a relatively new build of Kali and a simple apt-get update and upgrade command, it should work fine here.Before you get started I recommend switching to root by issuing the command:sudo -i Step 1:apt-get update -y && apt-get upgrade -yNext,
Artificial Intelligence (AI) has transformed how we interact with technology. Large language models (LLMs) are among the most prominent advancements, with OpenAI's ChatGPT being a well-known example. However, in specialized environments such as Linux, particularly for development, scripting, and com