AI attack surface
7 min read
What is an attack surface in cybersecurity?An attack surface is every entry point a cybercriminal could exploit—devices, accounts, apps, and connections. See the seven types and how to reduce yours.
5 min read
What is AI detection and response (AIDR)?AI detection and response (AIDR) is Expel's emerging category for detecting AI-powered threats, agentic risk, and prompt attacks. See how it works.
5 min read
What is MITRE ATLAS? The framework for AI and ML threatsMITRE ATLAS maps how bad actors attack AI and machine learning systems. See the framework, its tactics, and how Expel labels detections against it.
4 min read
What is prompt poaching? How attackers steal AI promptsPrompt poaching is the theft or reverse-engineering of a proprietary AI prompt. Learn how it happens, why it matters, and how to protect your prompts.
4 min read
What is prompt security? A guide for teams building on AIPrompt security explained: how to protect AI prompts from injection, poaching, and misuse—plus how Expel's threat hunts cover it.
5 min read
What is shadow AI? Definition, risks, and examplesShadow AI means employees use AI tools without IT or security approval. Learn the risks, real examples, and how Expel spots it in customer environments.
4 min read
Agentic AI security solutions: What to look forEvaluating agentic AI security solutions? Get the 5 buying criteria, key vendor questions, and a comparison table before you commit to a platform.
7 min read
What are AI cyber attacks? How attackers use AI in 2026AI cyber attacks use AI to automate, speed up, and adapt an intrusion. See the main types of AI attacks in 2026, real examples, and how to defend.
6 min read
What is a prompt injection attack?A prompt injection attack hijacks AI systems by embedding malicious instructions in content they process. Learn how it works and how to defend against it.
7 min read
How does AI threat detection work?AI threat detection uses machine learning and behavioral analytics to identify both known and novel attacks in real time.
