July 24, 2015 at 6:58 am
All we know is that the brain mass - body mass ratio of a species implies the species intelligence and that an individual with an abnormally large brain isn’t necessarily smarter than the average person
.
Just as not every computer is x times more intelligent because of its core speed increase of a factor x. I know the difference between my new 4GHz computer and my first 4MHz computer running wfw 3.11 isn't a factor 1000. No way near. But I bet that any current chess program, even if it is not that good would beat a chess program from the beginning of the nineties[/quote]
Just as not every computer is x times more intelligent because of its core speed increase of a factor x. I know the difference between my new 4GHz computer and my first 4MHz computer running wfw 3.11 isn't a factor 1000. No way near. But I bet that any current chess program, even if it is not that good would beat a chess program from the beginning of the nineties
July 24, 2015 at 6:58 am
I guess there are narrowly defined domains where computers seem to be improving their capabilities
Heh - the way you phrased that it seems like you are referring to them as having their own capabilities, not something given to them by the maker of the computer.
Maybe Steve is really working for SkyNet and trying to divert us from the fact that it is coming. 😛
David
@SQLTentmaker“He is no fool who gives what he cannot keep to gain that which he cannot lose” - Jim Elliot
July 24, 2015 at 7:09 am
SkyNet could have been very buggy and not self aware. How many times have we confused a bug with an indented feature of the program?:Whistling:
July 24, 2015 at 7:46 am
Any sufficiently advanced technology is indistinguishable from magic. CLARKE 1973.
Let us add a corollary.
Any sufficiently advanced AI is indistinguishable from Natural Intelligence.
Given this will someday be true, AI is coming, but may take centuries to evolve. Given a lifespan of a decade or two at most, most systems will fail before obtaining anywhere near that capability. Just as shuttle voting circuits were outmoded but too complex to risk replacement, so too the complexity of AI once it passes the thresholds of indispensable usefulness. At that point, it will reach it's Peter Principle and have risen to it's limit of competence. Adding more cores and more storage will fail to increase its usefulness for immediate interventions, but a place can be reserved for deep thinking resolutions to unanswerable complexity. Think weather control, DNA analysis, drug formulations, etc
After all, do we not also regulate our elders to the same, sheltering them from the brash newness of modern life while probing their conscious thought for wisdom earned through long life? If you don't, you are missing one of humanities greatest resources. Go visit the old man in the cave at the top of the mountain, or ask the impossible question to the AI in the dungeon. You will be surprised at the simplicity of the answers thus revealed.
July 24, 2015 at 7:57 am
I'm not concerned about robot overlords exterminating mankind in the future, but I do believe that artificial intelligence has created a sort of soft tyranny here and now. Think about how consumer credit scoring impacts your life, how automated trading impacts the stock market, and how a web search can put your name on a terror watch list.
"Do not seek to follow in the footsteps of the wise. Instead, seek what they sought." - Matsuo Basho
July 24, 2015 at 7:59 am
Overpopulation depends on the marker for which you are comparing population:
Humans by no means populate a vast majority of the habitable surface area in which we can survive.
However, humans are consuming natural resources at a rate for which the planet can not sustain its existing and past levels.
For example, the amount of fossil fuels being consumed far surpasses the natural replenishment. This is proven by the statements of the energy companies who say it is getting harder and harder to find the supplies. There are other examples that could be stated as well, but I won't go into them for time's sake.
So in relation to the supplies of natural resources and sustaining their levels, we are overpopulated based on per person use. Given, not everyone uses the same amounts. However, the waste of industrialized nations (especially the United States) does have adverse effects on other parts of the planet...as much as we'd like to think it doesn't because it's not immediately obvious or significant in our eyes.
As for AI...I think when machines become self aware and aware of their place in the pecking order...and have a sense of need to achieve...then we're in trouble. Otherwise, me and the Bunn coffeemaker in the kitchen at work will be best buddies. 😀
July 24, 2015 at 8:55 am
The problem with technology is that at some point it stops being about science and practical commercial applications and instead is driven by pedestrian consumer demand and marketing. I mean, the only thing more pointless than an internet connected toaster is a self-aware toaster. Where is the innovation?
My refridgerator stopped working a few days back, and when I purchase a new one this weekend, it's going to be the most basic model with no sensors, USB port, and not even an ice maker. Why spend $2500 for a high-tech box to keep my food cool, when the $600 model will do the same job? I wish I could buy a 1990 model refridgerator, it would last longer and require less maintenance. Maybe a factory south of the border still makes these things, and I can buy it on the black market.
"Do not seek to follow in the footsteps of the wise. Instead, seek what they sought." - Matsuo Basho
July 24, 2015 at 9:13 am
Eric M Russell (7/24/2015)
The problem with technology is that at some point it stops being about science and practical commercial applications and instead is driven by pedestrian consumer demand and marketing. I mean, the only thing more pointless than an internet connected toaster is a self-aware toaster. Where is the innovation?My refridgerator stopped working a few days back, and when I purchase a new one this weekend, it's going to be the most basic model with no sensors, USB port, and not even an ice maker. Why spend $2500 for a high-tech box to keep my food cool, when the $600 model will do the same job? I wish I could buy a 1990 model refridgerator, it would last longer and require less maintenance. Maybe a factory south of the border still makes these things, and I can buy it on the black market.
Amen to the not buying "smart" devices/products. I avoid smart TVs and stuff like the plague. It's bad enough hackers get into my insurance company, medical provider, dad's retirement at OPM, et al. They don't need to hack my TV too.
BTW, check in your area for used appliance shops or repair guys who have them. I know here several years ago, I got a washer and dryer (late 90s models) for the pair for $250. They lasted me for about 10 years. When I went to Home Depot and bought a new pair, I got the cheapest things they had that were scratch and dent. Same with the fridge I got. Scratch and dented basic model...was a little over $250. I didn't care about dings...as long as it chilled my food...that's all that mattered.
As for the innovation...having an appliance that could tell when it needs cleaning rather than a schedule...that can actually check your calendar and request input from you about if dinner at 6pm is still good.
Holy heck...I just thought about something...I could have an AI wife...and never have to worry about alimony or child support lol 😎
July 24, 2015 at 9:38 am
Is Skynet coming?
Computers are only as smart as the people who program them.
So we have nothing to worry about.
😉
July 24, 2015 at 9:58 am
Provided the guys who build the AI system have read and taken on board Asimov's 3 laws of Robotics and considered the changes required as discovered by Giskard and Daneel Olivaw to implement the zero law "A robot may not harm humanity or through inaction allow humanity to come to harm".
An evolved machine intelligence may be very different to what we consider intelligent but intelligent none the less.
July 24, 2015 at 10:28 am
Do I think that IA will one day be a reality? Yes, limited as others have said but it will be real. Will it include the ability to creatively reason as in combining three or more facts and building a probability of success or a potential thesis? I believe it will. And some AI may already be able to do that now.
Will it be able to reason or solve a problem? As long as there is an algorithm for it, yes. The really simple stuff like when to insert line breaks in text are easily done.
And can a computer do stuff like read digital text and transform it to words? Yes, but not all the words will be pronounced correctly.
But can it write an original book or work of fiction? Not without a lot of help.
So there are needs we have for machines that operate using deductive reasoning, or pattern recognition and formulating some result, and some or many of those exist. We will see more and more robust ones in the future.
But there are limits to what a machine can do. So in summary, limited AI yes, complete package I really do not believe so.
M.
Not all gray hairs are Dinosaurs!
July 24, 2015 at 10:41 am
What makes AI so useful; it's not just the fact that it can make decisions faster, it's that an AI is (ideally) not subject to the same biases, physical shortcomings, and logical inconsistencies that limit human intelligence. Think about ants or bees; they have tiny and very primitive brains, but they are able to dedicate themselves to a task and collaborate with peers at a level that humans never could.
A perfect AI shouldn't attempt to emmulate human intelligence. Let's face it; we humans are inherently flawed; that's why we build computers to do things like monitor the core temperature of a nuclear reactor in the first place. Instead, it should emmulate the intelligence of an ant or a bee, only directed toward a task for which we define.
"Do not seek to follow in the footsteps of the wise. Instead, seek what they sought." - Matsuo Basho
July 24, 2015 at 10:48 am
Eric M Russell (7/24/2015)
I'm not concerned about robot overlords exterminating mankind in the future, but I do believe that artificial intelligence has created a sort of soft tyranny here and now. Think about how consumer credit scoring impacts your life, how automated trading impacts the stock market, and how a web search can put your name on a terror watch list.
The systems we have today, even IBM's Watson, are more of knowledge base systems than an actual AI, and something like this:
http://news.discovery.com/tech/robotics/self-aware-robot-solves-riddle-150724.htm
is considered a great accomplishment. Think about the times on Jeopardy (it was only a few but still) where the response required some kind of reasoning, common sense, or other ability greater than fact retrieval and Watson came up with odd answers that made the audience laugh. I view these kinds of systems more like a preschool aged child, capable of learning many things, being able to retrieve that knowledge, and even come up with some application of that knowledge, but not really understanding that knowledge or being able to reason with it beyond a rudimentary level.
Perhaps someday, but I think there's a long way to go before AI will reach that level.
July 24, 2015 at 10:54 am
anacap (7/24/2015)
The destruction of all jobs performed by humans is an economic fallacy. Automation is sought after to reduce costs in search for higher margins. Automation cannot reach a point where revenue (sales) is destroyed because no one is able to buy things because they all lost their jobs. Automation will destroy some jobs but in the process create new ones.
Your first sentence is a bit extreme. There's a long way between the early 20th century, with the boom of manufacturing and then destruction of all jobs. We do see many fewer people needed in industries as a whole as automation and computing efficiencies take over. Certainly as the US and other developed countries grow their infrastructure, mid pay, mid skilled jobs are needed. We can use computers to reduce labor.
There will be new opportunities and new jobs created. For sure. However is there a net gain or loss, and perhaps more importantly, are the skills required for newer jobs available widely to the large populace?
July 24, 2015 at 10:55 am
GeorgeCopeland (7/24/2015)
Overpopulation is a myth. However I do agree that some countries and areas are overpopulated for how they live and the technology they use. For instance California has a water shortage but there are multiple technical solutions to that problem but until the problem is solved California could be deemed as overpopulated. Thus it also depends on the observers perspective.
The most densely populated place on the planet is downtown Tokyo. From where I am right now, you could walk a few miles away and not see a single human structure. Wherever there are more people, there is more food, this is because humans produce it. Obviously, we are nowhere near the population capacity of our planet.
Obviously? I don't think we are, but do we know. Do we have data on this? This is one place where we want to be studying and working to ensure we don't have issues. I think it's easy to be extreme on this issue (either way).
We'll adapt, but will we do so well, or poorly?
Viewing 15 posts - 16 through 30 (of 68 total)
You must be logged in to reply to this topic. Login to reply