Can you teach ethics to algorithms?

When people think of the ethical issues surrounding algorithms and AI, too many of us think of killer robots or movies like "The Matrix." But plenty of reasonable people are now rightly concerned that algorithms, far from being unbiased, can be used to perpetuate unjust or racist results. Last May, ProPublica published an article declaring that, "There's software used across the country to predict future criminals. And it's biased against blacks." ProPublica went on to detail how an algorithm used by parole boards to predict whether a criminal would re-offend was more likely to give bad scores to blacks than whites. There are plenty of other examples of algorithms which crank out ethically problematic results.