3 ms·
Agreed, but it's not like it's OK to say "Well sure, so far we have a kill rate of 50x a human driver, but why don't we wait until Uber kills at least 20 people
by drcode 9y ago
Agreed, but it's not like it's OK to say "Well sure, so far we have a kill rate of 50x a human driver, but why don't we wait until Uber kills at least 20 people first so we can determine proper statistical significance"
- jstanley 9y agoWhy not? Why in this case is it acceptable to make decisions on non-statistically significant data? You would exercise more rigour than that in a throwaway A/B test.
- jazoom 9y agoSame reason clinical trials stop early if someone dies. It might just be bad luck. Unfortunately we'll never know and that drug might never be tested again. It might have been an amazing drug. In this case, based on how it played out, I personally wouldn't want Uber's self-driving cars near me.
- tedsanders 9y agoIt is statistically significant, at a 0.03 level: https://news.ycombinator.com/item?id=16655081 https://news.ycombinator.com/item?id=16655081
- jstanley 9y agoExcellent