Amine Raji, PhD

AI security research on agents in production. Every claim reproduced in a lab first.

HOME
ABOUT
SPEAKING
NEWSLETTER
RESOURCES

AI Security Insights

Weekly digest of AI security threats, LLM vulnerabilities, and what security teams actually need to know.

Subscribe — it's free

Home Categories Architecture

Category

Architecture 2

The Missing Control: A Design Proposal for Tool-Output Redaction in AI Agents Aug 29, 2026
The Containment Ladder: Four Rungs That Keep an AI Agent's Mistakes Small Aug 1, 2026

Recently Updated

Cosine Similarity Is Not a Safety Property
The Injection That Passes Every Filter
The Missing Control: A Design Proposal for Tool-Output Redaction in AI Agents
Agent Identity: What the Standards Fix, and What They Leave Open
The Containment Ladder: Four Rungs That Keep an AI Agent's Mistakes Small

Trending Tags

AI Security Agentic AI MCP Prompt Injection LLM OWASP Kubernetes Red Team Threat Modeling Tool Poisoning

© 2026 Amine Raji, PhD. Some rights reserved.

Contact · Privacy

Using the Chirpy theme for Jekyll.

Trending Tags

AI Security Agentic AI MCP Prompt Injection LLM OWASP Kubernetes Red Team Threat Modeling Tool Poisoning

A new version of content is available.