Summer Sale · 25% off · SUMMER25
8kSec

Interactive Learning Roadmap

How to Learn AI Security

A dependency-ordered path from LLM fundamentals to model-level exploitation — across prompt injection, jailbreaking, RAG, AI agents and MCP, and the deeper ML-security layer of backdoors, extraction and adversarial machine learning. Start on the shared foundation, then fork into the specialization you want. Every node explains what it is, why it matters, and what it takes to master.

Everything here is testable on models you can run yourself. Practice on local models (Ollama / Hugging Face), deliberately-vulnerable LLM apps, and the free 8kSec Battlegrounds AI challenges — and treat every technique as authorized-testing-only, against systems you own or have permission to test.
Beginner Intermediate Advanced Offensive Defensive Fully detailed

Tip: click any node for a full checklist of the skills, knowledge and tools it takes to master it — plus an honest note where a skill only comes from research or paid training · filter by track or offense/defense · Esc closes.

01–05

Foundations

The shared trunk — five milestones every AI security practitioner shares before specializing, sequenced by dependency. Each opens a full checklist of the skills, knowledge and tools it takes to master it.

After the foundations You can reason about how an AI system is built and where it can be attacked — the toolkit of a junior AI security engineer / AI red-team associate. The specific attack and defense classes come next. Real roles also expect a portfolio of write-ups and hands-on labs.
the path forks

The application-security track

LLM Application & Agent Security

Attacking and defending the AI application itself — prompt injection, jailbreaks, RAG, agents, tools and MCP. Where most real-world AI risk lives today; no ML PhD required to start.

Prompt Injection & Jailbreaking

Direct injection
Jailbreaking
Milestone master everything above and you have the skills for the LLM Application Pentester role — with a portfolio and enough hands-on practice
Automated & advanced
Milestone master everything above and you have the skills for the AI Red Teamer (LLM) role — with a portfolio and enough hands-on practice

RAG & Data Security

RAG fundamentals
Poisoning & retrieval attacks
Milestone master everything above and you have the skills for the AI Application Security Engineer role — with a portfolio and enough hands-on practice
Data leakage

Agents, Tools & MCP

Agent fundamentals
MCP & multi-agent
Milestone master everything above and you have the skills for the AI Agent Security Specialist role — with a portfolio and enough hands-on practice
Output handling

LLM App Defense

Guardrails
Secure design
Milestone master everything above and you have the skills for the AI Security Engineer (Defensive) role — with a portfolio and enough hands-on practice

The ML-security track

Model & ML Security

Going below the prompt — the model and the ML pipeline itself. Adversarial machine learning, training-data extraction, poisoning, backdoors, model theft, and the defenses, red-teaming and governance around them. The deep end; math- and ML-heavy.

Model-Level Attacks

Adversarial ML foundations
Privacy attacks
Milestone master everything above and you have the skills for the AI/ML Security Researcher role — with a portfolio and enough hands-on practice
Backdoors & supply chain
Milestone master everything above and you have the skills for the AI Model Security Researcher role — with a portfolio and enough hands-on practice

ML Defense, Red Teaming & Governance

Defense & robustness
Red teaming & tooling
Milestone master everything above and you have the skills for the AI Red Team Engineer role — with a portfolio and enough hands-on practice
Incident response & governance
Milestone master everything above and you have the skills for the AI Security & Governance Lead role — with a portfolio and enough hands-on practice

A free, open study plan for learning AI & LLM security. Every node points to the standards, tools and research on the topic — plus, where relevant, the hands-on 8kSec course that goes deeper.

8ksec.io/roadmaps/ai-security · a free resource for the AI security community