Analysis
CyberTalks: Data Poisoning Attacks on ML & Agentic AI Systems | Jason Ross |COASP - YouTube
Publication date not yet evaluated · Added · 4 reports youtube.com
Page published
Coverage timeline
Why it matters
Data poisoning of training sets, fine-tuning data, and RAG knowledge bases lets attackers implant hidden, validation-surviving backdoors into AI and agentic systems, a defense-relevant attack surface that widens as knowledge bases update faster than model weights.
A CyberTalks/COASP presentation and accompanying explainer describe AI data poisoning — attacks that corrupt the data an ML or agentic AI system learns from, fine-tunes on, or retrieves so the model behaves as an attacker intends. The material covers training-data, fine-tuning, alignment, and RAG/vector-database poisoning, distinguishes it from prompt injection and model poisoning, and cites research (e.g. Carlini et al.'s 'Poisoning Web-Scale Training Datasets is Practical') showing that under 1% of poisoned data can change behavior.