Skip to main content

Become the AI security hire every org needs

Hands-on AI security training, from your first prompt injection to a certification that proves you can do the job.

Start learning for freeExplore AI training
Upskilling 9+ million learners and teams worldwide
GoogleKPMGAccentureHuntressSS&CThreatLocker
Cypheron CTF challenge scoreboard.Detecting Adversarial Attacks training room.AI agent terminal with real-time security logs.

AI Security at TryHackMe

Everything you need to go from your first lab
to AI security certified

Follow the guided path from the fundamentals to job-ready, prove what you can do with
the AI1 certification, and drop into free hands-on rooms whenever you want to go deeper.

Explore more
AI Threat Modelling room thumbnail.
room

AI Threat Modelling

Assess and mitigate enterprise AI/ML risks via systematic, defender-focused auditing.

medium

Walkthrough
AI System Reconnaissance room thumbnail.
room

AI System Reconnaissance

Discover AI infrastructure by scanning ML services, frameworks, and extracting metadata from APIs.

medium

Walkthrough
AI Security Threats room thumbnail.
room

AI Security Threats

Explore the vulnerabilities AI introduces, how attackers exploit it, and how defenders fight back.

easy

Walkthrough
AI Models & Data room thumbnail.
room

AI Models & Data

Explore how data is fundamental to AI security, and the models which power it.

medium

Walkthrough
Learning path
aisecurity

AI Security

AI is the new attack surface. Learn to break it, defend it, and secure it from the ground up.

  • Understand how AI systems are built and deployed
  • Attack and defend LLMs with prompt injection and jailbreaking
  • Secure AI supply chains and training data
  • Exploit OWASP LLM vulnerabilities against real agents

intermediate

Certification
ai-security

AI Security Level 1 (AI1)

Attack and defend real AI systems across 13 hands-on scenarios. Your entry point into the fastest-growing specialisation in cyber security.

AI Fundamentals module thumbnail.
module

AI Fundamentals

This module introduces learners to the foundational concepts of artificial intelligence and machine learning, establishing the mental models needed before diving into security topics. Rooms cover AI/ML threat categories, how models consume and process data, prompt engineering basics, and how AI can be used to augment existing roles using digital forensics as an example with a companion challenge to close off.

medium

Secure AI Systems module thumbnail.
module

Secure AI Systems

This module explores AI systems as an attack surface, covering secure architectural design principles and how to identify weaknesses at the system integration layer. Learners examine LLM-specific security concerns, apply threat modelling frameworks like STRIDE and OWASP in AI contexts, and practice attack surface discovery when AI components are present. The module concludes with a static site exercise where learners put their AI threat modelling skills to the test.

medium

2026: An AI Odyssey module thumbnail.
module

2026: An AI Odyssey

Module containing all event rooms for the "2026: An AI Odyssey" CTF

medium

Defending AI module thumbnail.
module

Defending AI

This module explores how adversaries exploit machine learning models using adversarial inputs, data poisoning, and evasion techniques that bypass traditional defenses. You’ll start by identifying these attacks and learning how they impact model integrity. Then, you’ll implement defensive strategies like adversarial training and input validation to harden your systems. Finally, you’ll leverage AI itself to assist in blue team operations by automating triage, detecting anomalies, and accelerating forensic investigations. By the end, you’ll know how to secure ML pipelines and weaponise AI for defensive advantage.

Learning path
agentic-ai-security

Agentic AI Security

AI agents plan, act, and delegate on their own. Learn to hijack them, then lock them down.
  • Understand how agents plan, use tools, and delegate
  • Hijack agent goals and abuse tools and identities
  • Exploit agent code execution, memory, and communication
  • Work through the OWASP Top 10 for Agentic Applications (ASI01–ASI10)

intermediate

AI Forensics room thumbnail.
room

AI Forensics

Explore AI DFIR and learn how it boosts your investigation capabilities.

medium

Walkthrough
AI Supply Chain Security module thumbnail.
module

AI Supply Chain Security

This module focuses on the security of AI models from procurement through deployment. Learners investigate how supply chains are targeted (OWASP LLM03), how to detect tampered or untrusted artefacts at ingestion, and how embedded triggers, poisoned checkpoints, and compromised fine-tunes manifest in real-world scenarios. Two hands-on challenge rooms reinforce defensive skills in identifying and responding to malicious models.

medium

AI Threat Modelling room thumbnail.
room

AI Threat Modelling

Assess and mitigate enterprise AI/ML risks via systematic, defender-focused auditing.

medium

Walkthrough
AI System Reconnaissance room thumbnail.
room

AI System Reconnaissance

Discover AI infrastructure by scanning ML services, frameworks, and extracting metadata from APIs.

medium

Walkthrough
AI Security Threats room thumbnail.
room

AI Security Threats

Explore the vulnerabilities AI introduces, how attackers exploit it, and how defenders fight back.

easy

Walkthrough
AI Models & Data room thumbnail.
room

AI Models & Data

Explore how data is fundamental to AI security, and the models which power it.

medium

Walkthrough
Learning path
aisecurity

AI Security

AI is the new attack surface. Learn to break it, defend it, and secure it from the ground up.

  • Understand how AI systems are built and deployed
  • Attack and defend LLMs with prompt injection and jailbreaking
  • Secure AI supply chains and training data
  • Exploit OWASP LLM vulnerabilities against real agents

intermediate

Certification
ai-security

AI Security Level 1 (AI1)

Attack and defend real AI systems across 13 hands-on scenarios. Your entry point into the fastest-growing specialisation in cyber security.

AI Fundamentals module thumbnail.
module

AI Fundamentals

This module introduces learners to the foundational concepts of artificial intelligence and machine learning, establishing the mental models needed before diving into security topics. Rooms cover AI/ML threat categories, how models consume and process data, prompt engineering basics, and how AI can be used to augment existing roles using digital forensics as an example with a companion challenge to close off.

medium

Secure AI Systems module thumbnail.
module

Secure AI Systems

This module explores AI systems as an attack surface, covering secure architectural design principles and how to identify weaknesses at the system integration layer. Learners examine LLM-specific security concerns, apply threat modelling frameworks like STRIDE and OWASP in AI contexts, and practice attack surface discovery when AI components are present. The module concludes with a static site exercise where learners put their AI threat modelling skills to the test.

medium

Learning path
aisecurity

AI Security

AI is the new attack surface. Learn to break it, defend it, and secure it from the ground up.

  • Understand how AI systems are built and deployed
  • Attack and defend LLMs with prompt injection and jailbreaking
  • Secure AI supply chains and training data
  • Exploit OWASP LLM vulnerabilities against real agents

intermediate

Certification
ai-security

AI Security Level 1 (AI1)

Attack and defend real AI systems across 13 hands-on scenarios. Your entry point into the fastest-growing specialisation in cyber security.

AI Fundamentals module thumbnail.
module

AI Fundamentals

This module introduces learners to the foundational concepts of artificial intelligence and machine learning, establishing the mental models needed before diving into security topics. Rooms cover AI/ML threat categories, how models consume and process data, prompt engineering basics, and how AI can be used to augment existing roles using digital forensics as an example with a companion challenge to close off.

medium

Secure AI Systems module thumbnail.
module

Secure AI Systems

This module explores AI systems as an attack surface, covering secure architectural design principles and how to identify weaknesses at the system integration layer. Learners examine LLM-specific security concerns, apply threat modelling frameworks like STRIDE and OWASP in AI contexts, and practice attack surface discovery when AI components are present. The module concludes with a static site exercise where learners put their AI threat modelling skills to the test.

medium

2026: An AI Odyssey module thumbnail.
module

2026: An AI Odyssey

Module containing all event rooms for the "2026: An AI Odyssey" CTF

medium

Defending AI module thumbnail.
module

Defending AI

This module explores how adversaries exploit machine learning models using adversarial inputs, data poisoning, and evasion techniques that bypass traditional defenses. You’ll start by identifying these attacks and learning how they impact model integrity. Then, you’ll implement defensive strategies like adversarial training and input validation to harden your systems. Finally, you’ll leverage AI itself to assist in blue team operations by automating triage, detecting anomalies, and accelerating forensic investigations. By the end, you’ll know how to secure ML pipelines and weaponise AI for defensive advantage.

Learning path
agentic-ai-security

Agentic AI Security

AI agents plan, act, and delegate on their own. Learn to hijack them, then lock them down.
  • Understand how agents plan, use tools, and delegate
  • Hijack agent goals and abuse tools and identities
  • Exploit agent code execution, memory, and communication
  • Work through the OWASP Top 10 for Agentic Applications (ASI01–ASI10)

intermediate

AI Forensics room thumbnail.
room

AI Forensics

Explore AI DFIR and learn how it boosts your investigation capabilities.

medium

Walkthrough
AI Supply Chain Security module thumbnail.
module

AI Supply Chain Security

This module focuses on the security of AI models from procurement through deployment. Learners investigate how supply chains are targeted (OWASP LLM03), how to detect tampered or untrusted artefacts at ingestion, and how embedded triggers, poisoned checkpoints, and compromised fine-tunes manifest in real-world scenarios. Two hands-on challenge rooms reinforce defensive skills in identifying and responding to malicious models.

medium

AI Threat Modelling room thumbnail.
room

AI Threat Modelling

Assess and mitigate enterprise AI/ML risks via systematic, defender-focused auditing.

medium

Walkthrough
AI System Reconnaissance room thumbnail.
room

AI System Reconnaissance

Discover AI infrastructure by scanning ML services, frameworks, and extracting metadata from APIs.

medium

Walkthrough
AI Security Threats room thumbnail.
room

AI Security Threats

Explore the vulnerabilities AI introduces, how attackers exploit it, and how defenders fight back.

easy

Walkthrough
AI Models & Data room thumbnail.
room

AI Models & Data

Explore how data is fundamental to AI security, and the models which power it.

medium

Walkthrough

Agentic AI Security Path · Included withMax plan.

Meet the next frontier: Agentic AI Security Path

Our purpose-built agentic platform puts real AI agents in your hands, each with its own tools and credentials. Hack them, defend them, and master the full OWASP Agentic Top 10.

Start learning

Why learn at TryHackMe

Learn AI security by breaking and defending real
systems, not reading about them

Companies are putting AI into production faster than they can secure it, and the people who can
are in short supply. You'll train to become one of them, on live models and real attacks.

25 lessons

A guided path across the AI attack surface.

Purpose-built AI platform

Break real models, live, in your browser.

Offense and defense

Attack AI systems, then build the defenses that stop you.

OWASP LLM Top 10

Mapped to the industry-standard framework.

Prove it · AI Security (AI1) Certification

Make your AI security skills official.

Attack and defend real AI systems across 13 hands-on scenarios. Practise prompt injection,then build the defense. Your way into the fastest-growing specialism in security.

Get AI certified
TryHackMe Certified AI1 badge.

Who's it for?

Know the security basics? You're ready for this.

No AI background needed - just bring your existing security skills, and we'll take you the rest of the way.

New To AI Security

You've heard about AI threats. Time to go hands-on.

Security Professionals

AI is landing in your org. Be the one who can secure it.

Developers Building With AI

You ship AI features. Learn how they break before someone else does.

Students & Career Builders

AI security roles are opening now. Show up with proof.

Your Edge

Start free. Get certified. Get hired.

Train on real AI systems, prove what you can do with
AI1, and stand out for the roles opening up right now.

Start learning for freeExplore AI training
Google logoKPMG logoAccenture logoHuntress logoThreatLocker logo