> the perception of time under the adrenaline rush of an imminent collision may be misleading. This may be one of the more ridiculous excuses I've seen for sel

> the perception of time under the adrenaline rush of an imminent collision may be misleading. This may be one of the more ridiculous excuses I've seen for self driving vehicle "screwups". Is the argument actually computers can't take action instantly compared to a human? If so, then self driving cars may as well not exist. It's well within a computer's capability to take evasive action for an accident like this far before a human driver - assuming sufficient sensor input. This is exactly the reason I want to see self driving tech rapidly accelerate - a car that has it's situational awareness 100% at all times, and thus can react instantly in a safe manner to any situation. I understand the software is not up to that point yet - and your response is exactly what I fear the most. I see endless DoS attacks against self driving cars once they get even a remotely substantial percentage of the market. If all programming reverts to "failsafe stop when things may be remotely potentially unsafe" we are going to see traffic problems of a proportion we have never experienced.