Max Andreacchi - Corgi Corp - Adversarial AI Research

Context Is Not Identity: Why AI Security is an Authorization Problem

AI security isn’t a prompt problem; it’s an authorization problem. Learn how identity, capability, and enforcement secure agentic AI systems.

Writeups

[HTB] Aero (Medium) - Walkthrough

Aero is a retired medium-difficulty box available on Hack The Box Labs. It is a Windows 11 system vulnerable to not one, but two

Code

> /code

Tools CrescendoAttacker Description: Multi-prompt testing tool that applies the crescendo technique to LLMs, progressively escalating prompts to probe for behavioral drift, policy boundary erosion,

Research

Emergent Architectural Leakage in Frontier Models: The Dual-Claude Phenomenon

TL;DR: A pleasant evening conversation last summer with Claude resulted in a possible disclosure of its internal architecture. Introduction As a red teamer,

Research

I Built a Robot to Argue With Other Robots (And It's Going Great)

I've spent the last several weeks personally testing AI model safeguards and becoming familiar with the nuance required when you're

Research

Tokenizing the Sandwich Debate: How NLP Models Weigh In on Hot Dogs

At this point in time, many of us have interacted with a chat-based AI agent and possibly been astounded by its ability to comprehend