Artificial Intelligence: Past and Future(cacm.acm.org)
cacm.acm.org
Artificial Intelligence: Past and Future
http://cacm.acm.org/magazines/2012/1/144824-artificial-intelligence-past-and-future/fulltext
20 comments
> This is like so many articles and books I've read (several mentioned in this article itself). Ok, AI is a threat to jobs up and down the ladder. I get it. Now what? What are we supposed to do about it?
Did you read "Homo Deus : A Brief History of Tomorrow" [1]? It seems to me this book tries to see things a bit further than most do. Basically the author says humanity will try to achieve happiness, immortality and power. If artificial intelligence can be controlled, then it will be used to reach these goals. The quest for happiness and power will then push us into modifying ourselves, to the point we'll turn into either machines or into a completely different species. Either way, that will indeed relegate human beings into a endangered species.
I personally believe there is nothing we can do, because we deeply want happiness, immortality and power, and we can't get them without modifying ourselves into something else.
1. http://en.wikipedia.org/wiki/Homo_Deus:_A_Brief_History_of_T...
Did you read "Homo Deus : A Brief History of Tomorrow" [1]? It seems to me this book tries to see things a bit further than most do. Basically the author says humanity will try to achieve happiness, immortality and power. If artificial intelligence can be controlled, then it will be used to reach these goals. The quest for happiness and power will then push us into modifying ourselves, to the point we'll turn into either machines or into a completely different species. Either way, that will indeed relegate human beings into a endangered species.
I personally believe there is nothing we can do, because we deeply want happiness, immortality and power, and we can't get them without modifying ourselves into something else.
1. http://en.wikipedia.org/wiki/Homo_Deus:_A_Brief_History_of_T...
"we'll turn into either machines or into a completely different species. Either way, that will indeed relegate human beings into a endangered species.
I personally believe there is nothing we can do, "
Why should we do something about in the first place? Evolution meant allways changing, so if we at some point change into a new species, why not? I mean, sure there will be struggles and fanatics will want to conserve the present human and fight everyone who became different, but they usually have no rational reason for their behavior. In other words, given how stupid so many humans are, I very welcome a genetic/cyborg improvement to us ...
Why should we do something about in the first place? Evolution meant allways changing, so if we at some point change into a new species, why not? I mean, sure there will be struggles and fanatics will want to conserve the present human and fight everyone who became different, but they usually have no rational reason for their behavior. In other words, given how stupid so many humans are, I very welcome a genetic/cyborg improvement to us ...
It's not so much as "should we?". I mean I don't want to put any moral assessment here. The truth is we don't quite know what will happen. Some of the scenarios are great, others are ugly or sad. Like for instance the idea that to reach happiness, mankind could use biochemical ways. To reach immortality and thus safety, we could chose claustration. That's the caricatural scenario where to make us "safe and happy", AI would lock us in jails with an abundant supply of heroin. A similar one is the one described in "The Metamorphosis of Prime Intellect"[1], which I like because it reminds me of a famous experiment that I find fascinating[2]
We don't know if changing ourselves would be for the better or the worse, but it looks that it will happen anyway and that will be consistent with the prediction that AI would cause the disappearance of Homo Sapiens, or at least its relegation to the status currently occupied by other mammals.
1. http://www.localroger.com/prime-intellect/
2. http://boingboing.net/2008/09/16/brain-implant-result.html)
We don't know if changing ourselves would be for the better or the worse, but it looks that it will happen anyway and that will be consistent with the prediction that AI would cause the disappearance of Homo Sapiens, or at least its relegation to the status currently occupied by other mammals.
1. http://www.localroger.com/prime-intellect/
2. http://boingboing.net/2008/09/16/brain-implant-result.html)
And then we may find ourselves right back at our oldest problem: philosophy - IMO one of the defining traits of our species (which AI must also take on), perhaps trying to decide which to choose from: happiness, immortality, or power?
> Ok, AI is a threat to jobs up and down the ladder. I get it. Now what? What are we supposed to do about it?
Make sure it doesn't fall exclusively in the hands of the 0.1%. We need to put the means of production in the hands of the people.
I envision a self-sustaining, self-reliant technology stack using solar energy, 3d-printing and automation, owned by a community or state, providing everything needed for its people. Maybe 1000 people could bootstrap a self sufficient community, if organized well.
Make sure it doesn't fall exclusively in the hands of the 0.1%. We need to put the means of production in the hands of the people.
I envision a self-sustaining, self-reliant technology stack using solar energy, 3d-printing and automation, owned by a community or state, providing everything needed for its people. Maybe 1000 people could bootstrap a self sufficient community, if organized well.
The closest model to what you're describing is Feudalism, and I don't mean say this in a dismissive way. I too believe that notions of Houses might spring up again. In one House comprising of about 50 people, only 5 would do the work which generates wealth, and puts food on the table. The others would do art and other such non-wealth producing activities. The incentive of the wealth-producers is to be able to claim that their House invented this or that artwork, etc.
We'll have to deal with these things as they come, how it will play out specifically, along what sort of timeline, and how society will react is not predictable, there are too many moving parts.
> We need to put the means of production in the hands of the people.
They are in the hands of the people. Most means of production can be bought on markets that are open to the general public.
They are in the hands of the people. Most means of production can be bought on markets that are open to the general public.
Sure, if you have a few million dollars on hand. Few hundred thousand might start a small company in service sector.
this really is a great vision that is both plausible and achievable within our lifetimes
We can't even build an OS that isn't broken in significnt ways. What makes anyone think we can build a bug-free self-replicating global technology/manufacturing stack?
Here is what we are "supposed" to do:
You are supposed to be afraid because you are competing with ruthlessly efficient machines who may replace you more cheaply. Fear will get you to focus on the AI threat/utopia and not be bothered with minor ethical/efficiency issues.
You are supposed to accept lower wages because of the above.
You are supposed to accept reduced worker safety to keep workers competitive with machines.
You are supposed to invest money in companies marketing AI.
You are supposed to support fewer regulations for innovative AI companies. If this means a few pedestrians get run over by self driving cars, then that is a risk AI companies are willing to accept("Move Fast and Break Things"). Fewer regulations mean AI markets can keep trade secrets protected and not reduced to easily regulated commodities.
You're supposed to buy "smart" products. Smart tvs, smart cars, smart dishwashers, etc. It's the least you can do since dumb products are going away and the new smart products are priced at less than exorbitant rates because AI companies are making up the difference by selling their data about you. It's okay if you don't fully understand these products, that's why they're smart and you are smart for buying something smart. Dumb products are for those who are in a lower socioeconomic class than you and for tin-foil hat conspiracy theorists unless they are expensive(in that case they will have an appropriate expensive label to advertise their high quality: "organic", "vintage", "antique", etc).
You're supposed to not let other issues distract you from the coming AI revolution, especially immigration. AI companies need immigrant labor to program the self-programming AI machines and to provide a human interface to the machines for the machine's owners and users. In some cases at first, the human robots will need to serve as the main AI, until their pure electronic and metal replacements arrive. Cheap human robots can fill the gaps, but regardless of cost, these humans must be controlled rigidly, just like robots.
You are supposed to be afraid because you are competing with ruthlessly efficient machines who may replace you more cheaply. Fear will get you to focus on the AI threat/utopia and not be bothered with minor ethical/efficiency issues.
You are supposed to accept lower wages because of the above.
You are supposed to accept reduced worker safety to keep workers competitive with machines.
You are supposed to invest money in companies marketing AI.
You are supposed to support fewer regulations for innovative AI companies. If this means a few pedestrians get run over by self driving cars, then that is a risk AI companies are willing to accept("Move Fast and Break Things"). Fewer regulations mean AI markets can keep trade secrets protected and not reduced to easily regulated commodities.
You're supposed to buy "smart" products. Smart tvs, smart cars, smart dishwashers, etc. It's the least you can do since dumb products are going away and the new smart products are priced at less than exorbitant rates because AI companies are making up the difference by selling their data about you. It's okay if you don't fully understand these products, that's why they're smart and you are smart for buying something smart. Dumb products are for those who are in a lower socioeconomic class than you and for tin-foil hat conspiracy theorists unless they are expensive(in that case they will have an appropriate expensive label to advertise their high quality: "organic", "vintage", "antique", etc).
You're supposed to not let other issues distract you from the coming AI revolution, especially immigration. AI companies need immigrant labor to program the self-programming AI machines and to provide a human interface to the machines for the machine's owners and users. In some cases at first, the human robots will need to serve as the main AI, until their pure electronic and metal replacements arrive. Cheap human robots can fill the gaps, but regardless of cost, these humans must be controlled rigidly, just like robots.
> fear-mongering
I rather think it holds great promises
I rather think it holds great promises
Repetition is a very persuasive technique. The more someone hears a claim, the more inclined they will believe it or at least attribute credibility to it, independent of the claim's merits.
Regardless of what one thinks regarding AI displacing jobs, there's been a remarkable growth spurt of technology and that begs lots of uncertainty about the future. When things get uncertain, people will speculate more, push their opinions on others, etc.
Sorry I'm not answering your question directly, but hopefully it suggests why we might be seeing all these "what's going to happen with AI" articles. You can bet there will be lots of opinions coming soon on "what should we do about it", but these are harder to tease out.
Regardless of what one thinks regarding AI displacing jobs, there's been a remarkable growth spurt of technology and that begs lots of uncertainty about the future. When things get uncertain, people will speculate more, push their opinions on others, etc.
Sorry I'm not answering your question directly, but hopefully it suggests why we might be seeing all these "what's going to happen with AI" articles. You can bet there will be lots of opinions coming soon on "what should we do about it", but these are harder to tease out.
Everywhere I turn, people speak of AI. Often to say little, or less, as is the case here. The ACM in full splendor (I'm a member).
Hah, yes. CACM selection of research papers is usually very interesting, but the opinion pieces often feel like they were produced just to meet a deadline. (I'm a member as well.)
This 2012 article is very dated, esp considering alphaGo.
Honestly, what is the purpose of articles like this other than fear-mongering? It offers no constructive suggestions, and the only reason it's on the front page of HN is because "Artificial Intelligence" in the title.