Prototyping an explainable machine

As artificially intelligent machines get smarter, their complex algorithms are getting more opaque, resulting in a black box. We don't know how the algorithm works, how a specific decision is made. The gaps in knowledge have powerful implications. How can we as designers bridge this gap, bring transparency and trust in our relationship with machines? The discussion around explainable AI (XAI) has focused on the technical challenge of interpretability.