All the talk and worry about AI seems to forget a discussion about the contexts in which AI will be deployed in. Already, conventional algorithms are trusted i

All the talk and worry about AI seems to forget a discussion about the contexts in which AI will be deployed in. Already, conventional algorithms are trusted in society, sometimes beyond reasonable or safe levels. The ethical questions posed by ethical AI design and deployment are exactly the ones our present-day reliance on (possibly poorly-designed) algorithms pose. Before we worry about AI harming us by an improper inference, we can, in the present moment, already worry about automated systems harming us by improper inferences (e.g. improper financial models, Tesla / aircraft driving assistance); the discussion on how to limit and selectively deploy AI is no different than that for present-day algorithms.