Exactly a year ago - a month earlier, a month later, it does not matter - we frankly stated: once a car with artificial intelligence will kill a person . To our surprise, it did not take long to wait: the AI had already begun to kill people. And in this regard, it makes sense to raise the eternal questions again: who is to blame and what to do?
“... although death at the hands of robots and even from a self-driving car can sound scary, we have to remind ourselves that hundreds of thousands of people die in car accidents every day. Although we can find consolation in the fact that humans are currently driving each car, statistically it will never be safer than when robots take on the duties of drivers. And when the advantages of self-driving cars surpass the safety factors — including the efficiency of daily trips, our city planning, and price movements when updating entire industries — we will fear them no more than personal computers.
The real risk associated with the development of robotics and artificial intelligence lies in their unintended use. Just as there is a fine line between nuclear energy and nuclear weapons, these advances in robotics and AI promise unpredictable abuses: deadly military applications, expanding the methods of espionage and hacking of exactly the same complex systems. Therefore, we must be fully aware of these risks and are ready to resolve the issues associated with them, while not forgetting the benefits they can bring to humanity.
Gates, Musk and Hawking absolutely correctly encourage caution when moving forward. However, they see the threat a little bit there, since human intention remains a far more formidable danger than the machines it invents. Unimaginable technologies are at arm's length; once they could only be dreamed about, but the people were always the main enemy of the people ”.
So we wrote a year ago.
Since May 2016, Tesla's autopilot (Autopilot, as the company calls it) went badly: on May 7, 2016, a car drove through the Florida state highway and crashed into a wagon crossing the road at one of the intersections. The Tesla car tore off the roof, and it flew about another 30 meters before stopping. Driver Joshua Brown died as a result of the accident. On July 10, 2016, on Sunday night, the Model X crashed into a fence in Montana, and again through the fault of the autopilot. Fortunately, the driver and passenger are both fine. But it could be otherwise.
... the main enemy of the people has always been only the people themselves. The media picked up both cases and smashed Tesla apart, criticizing the autopilot’s early exit. The most interesting thing is that now riding without autopilot becomes more dangerous than with it. And this death is no more than a drop in the sea of the millions of deaths that occur throughout the world through the fault of people.
“But the reality is, if you look at the figures, that driving with the Tesla autopilot is MUCH safer than without it, or in a car where it is not,” wrote Peter Diamandis, responding to an unfair media response.
It turns out that people really are again the enemy themselves? And what will happen next? We will watch how artificial intelligence kills hundreds of people, anywhere, in any way: for the sake of pharmaceutical experiments; eliminates unsuccessful designer babies; kills some people for the sake of others; deprives the lives of criminals for the sake of saving lives that they otherwise could carry? And we will look at it as the salvation of mankind? So, yes. We will have to accept less evil in order to get rid of more.
The article is based on materials .
Comments
Post a Comment