> Without understanding edge case scenarios, we cannot make claims like these. Actually, this claim isn't based on any particulars of any scenario. It's simpl

> Without understanding edge case scenarios, we cannot make claims like these. Actually, this claim isn't based on any particulars of any scenario. It's simply a common human cognitive pattern. We will readily accept a familiar risk over an unfamiliar risk, even if the unfamiliar risk is less risky. We all readily jump in our cars every day despite the statistical fact that car-miles are quite deadly when compared with alternatives like bus-miles, train-miles, or plane-miles. We're also likely to underestimate the magnitude of a risk when we have some perceived sense of control. Most drivers believe they are above average - so the risk of accidents is undervalued (due to ego). On planes, the risk of harm is overvalued because passengers have no sense of control in failure scenarios. Like I said, we SHOULD understand how new tech can fail. We ALSO need to be careful about the biases with which we approach a new technology and aware of what we're expecting from it. We tend to overvalue the ways in which a new tech can introduce unfamiliar harm, while undervaluing the ways in which a new tech can remove familiar harm that's currently taken for granted (hence, undervalued). This presents policy/planning problems that can inhibit introduction of society-benefiting innovations. This does not mean that conservative approaches to deploying new tech are wrong - far from it: conservative adoption is responsible - it simply points out that our natural human biases tend to maintenance of the status quo. My point is that, quite simply, we need to be as realistic and comprehensive in our establishment of the benefits a new tech could introduce, as well as the risks, in order to have a balanced perspective of its overall value.