In AI, the ratio of attention on hypothetical, future, forms of harm to actual, current, realized forms of harm seems out of whack. Many of the hypothetical forms of harm, like AI "taking over", are based on highly questionable hypotheses about what technology that does not