Self-driving cars have been a popular topic over the past few years, with large companies such as Google, Tesla but also more traditional car manufacturers investing enormous amounts of money into this technology. And it is not hard to see why either, the potential disruption of this technology is enormous. Not only our personal lives but also entire industries could be disrupted. If the transport sector could go driverless it would lead to massive cost savings and efficiency gains.
So this all sounds really good so why aren’t we there yet? Now you might think because it’s a hard thing to develop, it is not easy to create an algorithm that can safely drive around. What else could be the reason that so many companies have been spending such large amounts of money for such a long time?
Well let me introduce you to Comma.ai, comma.ai is a company created by George Hotz. George Hotz is somewhat of a legendary genius, at the age of 17 he was the first one to ever jailbreak the first iPhone and his software to do this had more than 2.000.000 users. He has made some further waves and has actually led an interesting life so far, I would highly recommend to check him out but that’s beside the point for this post.
Comma.ai has developed a self-driving algorithm which can be easily plugged into every kind of car regardless of which brand or model. If we must believe George himself it is not even generally applicable but also better than Tesla’s ai!
Whether these bold claims are true or not, one thing is clear they have a really competent self-driving algorithm. And if a few guys can make this from their garage it beckons the question why haven’t we had driverless cars years ago?
For more information check out this video:
Dear Rick,
I think that asking this question is totally legitimate and can offer several insights. I would like to answer you telling you about a video that I happened seen yesterday on an online channel about new technologies. It showed how a Uber self-driving car killed a person on the street because it confused this person with an “inanimate object”. Interestingly enough, there was a driver in the car who was supposed to act in case of emergency and who did not realize what was happening in time. This terrible occurrence made me understand that, although humans make as much as, if not more, mistakes than technology, trust from people toward technologies is not to give for granted. Revolutionary technologies such as self-driving cars, which in the future will completely turn upside down our everyday life, will be used by people, and for this reason they must be trusted by people. Until that does not happen, we can have the best technologies ever, but they are not worth anything without people making use of them. This, in turn, requires trust, and trust takes huge amounts of time and many demonstrations to sink in for such disruptive changes.
Hello, Rick! Thank you for an interesting post.
I think, that although the technology behind the self-driving algorithm and the road situation recognition is not the bottleneck in the implementation of the self-driving cars. The regulation behind those are a major barrier for the self-driving cars to enter the global market. Also, what I believe is the key to implementing self-driven cars in our life is the ethics behind it. With a person behind a wheel, we always know who is responsible for any situation on the road, however with an algorithm it gets more complicated. Would you sit in a car, which can kill you (in theory of course)?