Label flipping attacks in data poisoning
How label flipping data poisoning works, where it fits in the AI attack surface, and a practical Python walkthrough building a clean baseline for the attack.
AI data attacks across the pipeline
Walk through AI data attacks stage by stage, from data poisoning at collection through storage tampering, processing manipulation, and online poisoning.
The AI data pipeline and its attack surface
Map the AI data pipeline stage by stage and learn where each creates attack surface, from data collection through to the retraining feedback loop.
Regulating LLM abuse attacks
How the US and EU regulate LLM abuse attacks, covering the Take It Down Act, NIST AI RMF, the EU Digital Services Act, and the EU AI Act risk framework.
Mitigating LLM abuse attacks
How to defend against LLM abuse attacks using model safeguards, deployment-level filtering with Google Model Armor and ShieldGemma, and content monitoring.
LLM abuse attacks
How LLMs enable abuse attacks including propaganda, AI-generated phishing, misinformation, and hate speech, with real-world cases and threat characteristics.
Mitigating insecure output in LLM applications
How to defend against insecure LLM output handling with context-specific encoding, access control enforcement, rendering layer defences, and sandboxing.
LLM hallucinations and their security impact
How LLM hallucinations create security risks from financial liability to supply chain attacks via slopsquatting, with mitigation strategies at every layer.
Exfiltration attacks through LLM output
How Markdown image rendering in LLM applications enables data exfiltration, covering the mechanism, indirect prompt injection delivery, and real-world CVEs.
Insecure function calling in LLM applications
How function calling in LLM applications creates code execution, excessive agency, and injection risks, with techniques for each vulnerability class.