3 ms·
Sure. But you are assuming a good way to estimate that chance. Say you base it on previous performance: in the last 5 years, twice as many people have died driv
by lazaroclapp 10y ago
Sure. But you are assuming a good way to estimate that chance. Say you base it on previous performance: in the last 5 years, twice as many people have died driving non-self-driving cars compared to those riding self-driving cars. Then, the very next day, a massive hack / cyberattack happens that causes 10 million fatal accidents...