Finally, someone is asking the questions we should've been asking from the start. Interpretability is key to building trust in AI systems. https://www.reddit.com/user/InfinityZeroFive