Data Poisoning
AI SecurityThis glossary entry explains Data Poisoning for AI governance and model risk programs. The sections below summarize what the term means in plain language, why chief AI officers and cross-functional committees track it, where teams often get confused, and—when you are signed in—how it shows up across major industries and in expectations tied to the EU AI Act and NIST AI RMF. Use related links at the end of the page to explore neighboring concepts without losing context.
What It Means
Data poisoning is when attackers deliberately corrupt the training data used to build AI models, similar to contaminating ingredients before cooking a meal. This malicious data causes the AI system to learn incorrect patterns and make wrong decisions or produce biased results that favor the attacker's goals.
Why Chief AI Officers Care
Data poisoning can undermine the reliability and trustworthiness of your AI systems, leading to poor business decisions, regulatory compliance failures, and damaged customer relationships. As the executive responsible for AI governance, you need robust data validation processes and supply chain security to protect against these attacks that could compromise your entire AI strategy.
Real-World Example
An attacker could inject fake product reviews into an e-commerce recommendation system's training data, causing the AI to consistently recommend inferior products from a competitor, ultimately driving customers away and reducing sales.
Common Confusion
Many executives think data poisoning only happens through external hackers, but it can also occur through insider threats, third-party data vendors, or even unintentional contamination from poorly managed data sources.
Industry-Specific Applications
See how this term applies to healthcare, finance, manufacturing, government, tech, and insurance.
Healthcare: In healthcare, data poisoning could occur when attackers inject false patient records, manipulated medical images, or in...
Finance: In finance, data poisoning poses significant risks to algorithmic trading systems, credit scoring models, and fraud dete...
Premium content locked
Includes:
- 6 industry-specific applications
- Relevant regulations by sector
- Real compliance scenarios
- Implementation guidance
Technical Definitions
Explore more glossary terms
Discuss This Term with Your AI Assistant
Ask how "Data Poisoning" applies to your specific use case and regulatory context.
Start Free Trial