Questions
Common questions.
What is Training Data Poisoning?
Training data poisoning is an attack where an adversary corrupts some of the data used to train or fine-tune an AI model, causing the model to develop specific biases, backdoors, or vulnerabilities.
How does Training Data Poisoning work?
It is a supply chain risk for agents built on custom fine-tuned models and for models that learn from continuously collected feedback.
Which terms are related to Training Data Poisoning?
Closely related concepts include AI Bill of Materials (AI BOM), Foundation Model, Model Factsheet, World Model. Each is defined in the Prefactor glossary.
Keep reading