3 ms·

Your argument is one I've heard many variations of. Ive never found it very convincing. Here's why: 20ish years ago, when I started my career, Google and Fac

Your argument is one I've heard many variations of. Ive never found it very convincing. Here's why: 20ish years ago, when I started my career, Google and Facebook didn't exist. Amazon sold only books and was considered a bold experiment. No AWS. People argued about whether linux would be ever be used in production. The iPhone didn't exist. Wifi was a newish spec, but most people getting online (which wasn't most people!) relied on modems, or isdn if you happened to live in certain places. There was no "internet of things" - heck "information superhighway" metaphors were finally starting to phase out, and the average American would still talk about The World Wide Web so awkwardly you could hear the capital letters. Self driving cars were total sci fi that could never happen because it's hard, plus who would you sue? (Plus cest la change, plus cest la meme chose) (I might be off by a year or two, but the point remains) I entered the profession about 20 years ago and have seen all of this change. At my age, I can expect to be in the industry for most of the NEXT 20 years. Before I even retire. Given that the rate of change accelerates - https://ddcvqwvagfyuw.cloudfront.net/content/uploads/2015/12/adoption_of_tech_no_title.jpg https://ddcvqwvagfyuw.cloudfront.net/content/uploads/2015/12... - we have no idea what to expect by then, but can expect it to be very different, all while we are still struggling with how to deal with the impact on society that faster travel and communication introduced from before I was even born. Do I think AI will exist and turn on humanity in that time? Can't say. AI predictions have been notoriously bad. (Usually overconfident, but also in the other direction. 20 years ago it was obvious computers would be able to beat any grandmaster in chess sometime in our lives, but go was considered to be something dramatically past that. So I don't know if it will happen or even be able to happen. But I'm not convinced by anyone pointing out failures in current tech. That's like saying human flight can't happen because a lot of people failed at it, or that cars won't replace horses because they break down too much. Tech changes and it is change at a growing rate. To cite a metaphor: if you have a pond. A lily pad gets there. Next day you see there are 2. 4 the next day. Will they overgrow the pond? If you wait until the pond is half covered to worry about it (e.g. 20 days), you have only 1 day to handle it before 'maybe' is proven 'yes', not 20.

oldandtired 9y ago
You have been in the industry 20 years, I have been in the industry nearly twice that and as far as AI is concerned, it is a no show. Certainly we have systems that can play games, we have systems that automate factories and other processes, but none of these are AI. As humans have no understanding of intelligence and certainly not human intelligence, their ability to create an artificial one is effectively non-existent. Compared to the existing natural biological systems that we find around us, we cannot even duplicate their simplest processes with any aplomb. Our attempts are extremely inefficient in all areas, energy, speed, production purity, etc. If one looks around, one will see much change occurring. Yet, in so many ways, little changes. People remain people, the problems that plague people remain the same the same. If anything, our technology is only speeding up the rate at which these problems are occurring. Much of this is because we don't actually sit down and ask if we should be doing something at all. Just because we have varying success at developing various kinds of technology, doesn't mean we actually have the wisdom or understanding to use that technology properly or that we will achieve duplication of intelligence in a non-biological system. In regards to your pond analogy, instead of using exponential growth, I think we could say that it is logarithmic growth.

oldandtired 9y ago

You have been in the industry 20 years, I have been in the industry nearly twice that and as far as AI is concerned, it is a no show. Certainly we have systems that can play games, we have systems that automate factories and other processes, but none of these are AI. As humans have no understanding of intelligence and certainly not human intelligence, their ability to create an artificial one is effectively non-existent. Compared to the existing natural biological systems that we find around us, we cannot even duplicate their simplest processes with any aplomb. Our attempts are extremely inefficient in all areas, energy, speed, production purity, etc. If one looks around, one will see much change occurring. Yet, in so many ways, little changes. People remain people, the problems that plague people remain the same the same. If anything, our technology is only speeding up the rate at which these problems are occurring. Much of this is because we don't actually sit down and ask if we should be doing something at all. Just because we have varying success at developing various kinds of technology, doesn't mean we actually have the wisdom or understanding to use that technology properly or that we will achieve duplication of intelligence in a non-biological system. In regards to your pond analogy, instead of using exponential growth, I think we could say that it is logarithmic growth.