Free, high-quality AI security education — courses, live events, and resources for the people defending AI systems.
A full AI-security curriculum, taught by passionate instructors.
Weekly live discussions and monthly summits with expert speakers.
Taught by the people innovating AI security for a living.
Three hours, 1,200 registrants, an expert panel, five live security showcases, and new research on 31,000 MCP servers. Here's what you missed, and where to watch every session.
A four-stage model — Delegation, Deference, Dependency, Detachment — for how decision authority slips from accountable people to automated systems, drawn from failures from Knight Capital to Durban to the Boeing 737 MAX, and applied to AI agents and MCP in Africa.
From predictive-policing algorithms and AI risk scores to facial recognition and tools like Cybercheck and COMPAS, AI is quietly reshaping policing, courts, and criminal justice — sometimes solving cold cases in days, sometimes jailing the wrong person.
Your first stop in AI security — a tour of the technical and organizational forces shaping the field, from threat modeling to the frameworks now governing it.
An overview of MCP and its security implications, walking the OWASP MCP Top 10 through real-world incidents and the fixes that stop them.
The risks every team shipping LLMs needs to know, mapped to the OWASP LLM Top 10 and taught through real attacks.
This isn’t theoretical. I’m already applying it directly to how I think about the agents and MCP servers I work with day to day — permission scoping, what a tool is allowed to touch, what “authorized” actually means when an LLM is the one calling the shots.
Marc DavidStaff Security Engineer, Omada HealthFinish a course and get a shareable credential with its own public verification page.
Session recaps, new course drops, and the week's most important AI-security research — no noise.
