AI Glossary Term
What does “Black-Box Model” mean?
A model whose internal reasoning is difficult to interpret directly by humans.
How Black-Box Model works
Example
A deep neural network rejects a loan application. Its prediction depends on millions of numeric weights, so nobody can point to a simple rule explaining the decision. Teams then use extra tools, such as feature-importance methods, to get partial explanations.
Why it matters
When people cannot see why a model decided something, it is harder to check for errors or bias. In regulated areas such as credit and health, this limits where such models can be used.
Common questions
Are all AI models black boxes?
No. Simple models such as small decision trees or linear models are easier to read. Large neural networks are the harder cases.
Can black-box models be explained?
Partly. Post-hoc techniques estimate which inputs mattered, but these are approximations, not a full account of the model's reasoning.
Why use black-box models at all?
They often give higher accuracy on complex data such as images and language, so teams weigh accuracy against interpretability.