LIVE FEED
ATLAS

AI Attack Landscape

Adversarial techniques targeting AI/ML systems — mapped from real-world threat intelligence

MITRE ATLAS (Adversarial Threat Landscape for Artificial-Intelligence Systems) is a knowledge base of adversarial tactics and techniques targeting AI/ML systems, maintained by MITRE Corporation. Structured analogously to the well-known MITRE ATT&CK framework, it covers 101 techniques across 16 tactics — with expansions into agentic AI attacks, RAG-based attacks, and AI supply chain threats.

This page tracks how ATLAS techniques appear in our published threat intelligence. Use it to understand which attack vectors are active, which are emerging, and how they relate to each other.

→ Official MITRE ATLAS site

Time range:

Tactic Distribution

How attack activity distributes across the ATLAS tactic lifecycle

How to read: Each box represents one ATLAS tactic — a stage in the adversarial lifecycle. The number shows how many technique mentions appear in our articles for that stage. Click any tactic to expand it and see the individual techniques associated with that stage. Darker cells = more articles reference that technique. Click a technique cell for full details.

Emerging Techniques

Technique mention trajectory over the last 12 weeks

How to read: Each row shows one technique. The sparkline charts weekly mention counts — rising lines indicate growing attacker interest. The percentage on the right compares the last 4 weeks against the prior 4 weeks. Green ↑ = accelerating, grey ↓ = declining. Click any technique ID to see its full description.

Attack Patterns

Techniques that frequently appear together in the same article

How to read: Each node is a technique. Lines connect techniques that co-occur in the same article — thicker lines mean they appear together more often. This reveals composite attack chains: e.g., prompt injection (T0051) often co-occurs with data leakage (T0057), suggesting attackers chain these techniques. Larger nodes indicate more total mentions.

ATLAS → OWASP Mapping

How MITRE ATLAS techniques relate to OWASP LLM Top 10 categories

How to read: Each card shows an OWASP category and the ATLAS techniques that map to it. This helps you understand how granular attack techniques (ATLAS) roll up into broader vulnerability categories (OWASP) — useful for bridging operational threat intelligence with risk management frameworks.

◉ AI THREAT BRIEFING

Stay ahead of the threat.

Twice-weekly digest of critical AI security developments — every story mapped to MITRE ATLAS and OWASP LLM Top 10. Free.

No spam. Unsubscribe anytime.