Explained: How to tell if artificial intelligence is working the way we want it to, MIT News
![](https://news.mit.edu/sites/default/files/images/202207/MIT-Explanations-Explained-01-PRESS.jpg)
Deep-learning models have become very powerful, but that has come at the expense of transparency. As these models are used more widely, a new area of research has risen that focuses on creating and testing explanation methods that may shed some light on the inner-workings of these black-box models.
![](https://news.mit.edu/sites/default/files/styles/news_article__archive/public/images/202204/MIT-Shared-Interest-01-PRESS.jpg?itok=15Abo9k8)
Explained: How to tell if artificial intelligence is working the
![](https://news.mit.edu/sites/default/files/styles/news_article__archive/public/images/202205/MIT-Fair-Explanations-01-press.jpg?itok=6h8wOeWM)
Explained: How to tell if artificial intelligence is working the
![](https://news.mit.edu/sites/default/files/styles/news_article__archive/public/images/202205/MIT_Explanation-Summary-01-PRESS.jpg?itok=mbpegP7u)
Explained: How to tell if artificial intelligence is working the
![](https://news.mit.edu/sites/default/files/styles/news_article__archive/public/images/202201/MIT-Feature-Attribution-01-press.jpg?itok=3SBzJZBr)
Explained: How to tell if artificial intelligence is working the
![](https://news.mit.edu/sites/default/files/images/202207/MIT-Explanations-Explained-01-PRESS.jpg)
Explained: How to tell if artificial intelligence is working the