Posted originally in the old forum by sylvieskewsg:
For one second seems to be a dream. But in case if it comes true, then there are both advantages and disadvantages. But the question of advantages and disadvantages comes afterwards, first and the most important question is that is it really possible?
It is definitely possible. Google has put thousands of miles on its self driving car without human intervention. It is only a matter of time and which technologies will evolve fastest. Is there some specific reason you don’t think fully self driving cars are possible?
Let me correct that number of miles. As of May 2015 The Google self driving cars had 1.7 million miles on them, put on over a 7 year period. Here’s a nice overview about the Google technology.
No, self-driving cars will never be a reality. The technology is very impressive, but the very sophisticated nature of that technology produces the illusion of ultimate perfection, which can never be realized, yet is absolutely necessary. I do think it reasonable to expect such vehicles to perform with relative safety under normal highway conditions with clearly defined lane markers. But to be fully autonomous they would have to be able to negotiate the far more challenging conditions of city driving, not to mention highway conditions that are less than ideal, i.e., heavy rain, snow, sleet, fog, obscure or nonexistent lane markers, etc. Also, what happens when there IS an accident, how will such a vehicle be managed by law enforcement? Finally, as someone who’s spent many years programming computers, I can tell you there is NO 100% reliable way to assure that any program of any complexity is completely bug free. And the bug might show up only after thousands of hours of use. Also there is no way to assure that every computer chip in a system will last forever, or even be able to issue a warning in advance of failure.
Yes, it’s true that these systems will “learn” over time from past failures, but every new element added to any system will increase it’s complexity, which in turn increases the chances for something to go wrong.
There is of course too much money invested in automated vehicles to adequately regulate them, and for every accident we will be reminded of how much “safer” they are than human driven vehicles, so it seems inevitable that they will be put on the market. But after so many accidents no one will any more feel safe in them and since there has never really been a crying need for this technology, demand will vanish into thin air and tons of investor money will go down the drain.