Glossary
Special | A | B | C | D | E | F | G | H | I | J | K | L | M | N | O | P | Q | R | S | T | U | V | W | X | Y | Z | ALL
B |
|---|
BiasSystematic errors or distortions in the outputs of an AI system which reflect
or reenforce discriminatory patterns or stereotypes. This is often due to underlying
issues in the training data of an AI system (also see Prejudice). | |
Black box modelAn AI system whose internal workings are not easily interpretable or
transparent, even to its developers. | |
For further information, take a look at our frequently asked questions which may give you the support you need.
If you have any concerns about anything on this site please get in contact with us here.
