Back to Catalog
Owasp Llm
top10_2025
LLM04 - Data and Model Poisoning
Attackers manipulate training or fine-tuning data to compromise model integrity.
Intent & Description
'
🎯 Intent
Protect the integrity of data used for training, fine-tuning, and embedding from malicious manipulation.
📋 Context
Adversaries can inject malicious data into training sets, fine-tuning datasets, or embedding databases to alter model behavior, introduce biases, or create backdoors.
💡 Solution
Validate and sanitize all training data. Implement data provenance tracking. Use anomaly detection on training pipelines. Monitor model behavior for drift. Maintain clean reference datasets for comparison.'
Real-world Use Case
Use when collecting training data, fine-tuning models, or building embedding/vector databases.
Source
📌 TL;DR
Protect training data integrity. Validate data sources, track provenance, detect anomalies, monitor model behavior.
Advantages
- Maintains model reliability
- Prevents behavior manipulation
- Ensures data quality
- Protects against backdoors
Disadvantages
- Poisoned data can be subtle
- Detection is computationally expensive
- Requires clean baseline data