BT Futurologist On Smart Yogurt and the $7 PC
Posted by
kdawson
on Wed Sep 27, 2006 12:10 PM
from the i'll-be-back dept.
from the i'll-be-back dept.
WelshBint writes, "BT's futurologist, Ian Pearson, has been speaking to itwales.com. He has some scary predictions, including the real rise of the Terminator, smart yogurt, and the $7 PC." Ian Pearson is definitely a proponent of strong AI — along with, he estimates, 30%-40% of the AI community. He believes we will see the first computers as smart as people by 2015. As to smart yogurt — linkable electronics in bacteria such as E. Coli — he figures that means the end of security. "So how do you manage security in that sort of a world? I would say that there will not be any security from 2025 onwards."
Related Stories
[+]
Something in Your Food is Moving 378 comments
Dekortage writes "The New York Times has a report on probiotic food: food that has live bacteria in it. From the article: "[for Dannon's] Activia, a line of yogurt with special live bacteria that are marketed as aiding regularity, sales in United States stores have soared well past the $100 million mark.... Probiotics in food are part of a larger trend toward 'functional foods,' which stress their ability to deliver benefits that have traditionally been the realm of medicine or dietary supplements.""
This discussion has been archived.
No new comments can be posted.
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
Yogurt is already smarter than me (Score:5, Funny)
Re:Yogurt is already smarter than me (Score:5, Funny)
And if you were a hot chick, we'd demand proof.
Re:Yogurt is already smarter than me (Score:5, Funny)
I used to suffer from the same problem. Try opening the container from the other side.
Re: (Score:3)
See, that's why I'm an optimist about technology. Human intelligence should be able to keep one
Re:Yogurt is already smarter than me (Score:5, Funny)
Re: (Score:3, Funny)
That's it, I'm sticking with ice cream.
Re: (Score:3, Funny)
I tried this, and it just made things worse. First, I had to get a knife to cut through the bottom side, then all the yogurt fell onto the floor.
Re:Yogurt is already smarter than me (Score:5, Funny)
Futurologists... (Score:5, Insightful)
*ducks*
Right. (Score:5, Interesting)
Get a grip, for God's sake.
Re: (Score:3, Insightful)
Re:Right. (Score:4, Insightful)
That is the role of the "automatic" part in the modern phones
Smartitude: people vs computers (Score:5, Funny)
Computers as smart as "some" people im sure (Score:4, Insightful)
Re:Computers as smart as "some" people im sure (Score:5, Insightful)
When you have a computer that can beat you at chess without having a chess program installed, it's time to be concerned.
Re:Computers as smart as "some" people im sure (Score:4, Insightful)
Hmm. That doesn't explain why the chess program I wrote in college kicked my own ass every time I played against it
Re:Computers as smart as "some" people im sure (Score:5, Insightful)
Intelligence is like terrorism (or pornography), in that it's definable only with broad, nebulous, debatable borders. Chess is one kind of intelligence, and our current logic models are excellent here. Art is another kind of intelligence, and our current logic models are terrible here.
The problem with modern AI (and the flaw in Ian Pearson's predictions) is that we really don't understand many kinds and elements of intelligence. For instance:
- Spontaneous thought: Why do we think? What motivates us to keep thinking when we don't have a task to solve, or a logical process to follow?
- Associative memory: What element of our memory structure allows us to make prescient associations on the fly? Not just "green is a color, and so is blue," but "this song reminds me of one time when I was eating ice cream?"
- Creativity: Why are we good at coming up with surprising and unexpected insights? Modern AI tries this by billions and trillions of fumbling attempts to introduce randomness - but most of them are rubbish. But this is like evolution - which takes thousands or millions of years to innovate (randomly, clumsily) - and not like creative engineering.
- Emotion: We don't understand emotion at all. We've identified regions of the brain in which emotions occur, and particular hormones and hormone receptors that are involved. That's about it. The neuological basis of emotion remains a mystery.
These are just a few things that any human-competitive intelligence would need, but that we don't understand. Accordingly, it's completely impossible to predict when we will be able to model it, since we don't even understand it yet.Anyone who tells you differently is trying to sell you their book. ;)
- David Stein
Re:Computers as smart as "some" people im sure (Score:4, Insightful)
The other side of AI says that "my brain is magic, and I'm really smart and you can't possibly produce a robot as clever as me". I don't subscribe to that one - I think that's nonsense.
At minimum he's misrepresenting "the other side" of AI. As one professor (in the only college class on philosophy I've taken, btw) recounted, in the 60's a universal human language translator was inevitable and right around the corner. The problem is that these predictions were being made by technologists and not linguists -- people who didn't understand the problem. And language, on the surface, is a simple problem: languages have rules, exceptions to these rules and vocabulary that can be exhaustively enumerated -- a custom-fit problem for computers, right?
Turns out that machine translation from one language to another was a tad more complicated -- all due to a lack of understanding of linguistics. It's a problem for a linguist to solve, not a programmer or "AI Researcher".
We'll first understand how our minds work, and then we'll be able to create strong AI. A shrink can better tell us when this might happen than a technology futurist (and of course, there's plenty of good arguments that this will never happen).
IMHO, you're very right in pointing out that you run into basic problems once you start out trying to define what we mean by human intelligence; in fact, there's a very good argument to be made that when you start peeling away layers, a lot of what we understand as human intelligence is innately biological. As in, no strong AI in 100 years, no strong AI ever -- not created by human intelligence at any rate.
This doesn't seem to be a popular view of intelligence with technology-minded people: we seem to assume that the brain is the hardware and that our mind is the software -- so all you need is the right program running on your 386 and "poof" you have human intelligence.
That's how I felt myself actually; the prof's arguments didn't make sense to me until after a few years after I got my C in his class during an discussion of AI that I finally understood what he was saying all those years ago
Re: (Score:3, Insightful)
Re: (Score:3, Insightful)
I'm sure I'm not the first person to have thought of this, but what if we just simulated a biological human in software? Computers are already pretty good at simulating chemical reactions, physics, even as far as protein folding . . . why not learn to bui
Re: (Score:3, Interesting)
Not
$7 PC: Wrong (Score:5, Insightful)
Well, no one except Nintendo.
Re: (Score:3, Insightful)
Re: (Score:3, Insightful)
And flying cars and moonbases (Score:5, Insightful)
The problem is that they can only detect trends and can't really predict real things. So when you see a futurist going out on a limb and claiming that X is only 10 years away, they are hedging their bets that you will forget they ever made such a silly prediction 10 years from now. If they do manage to get something right, you can bet they'll be working overtime trying to get grants from RAND and MITRE for more futurism.
However, the reading of trends is a very important role of sociology. Only by accurately predicting what sorts of stresses and issues we will face in the near-term future can we sufficiently prepare ourselves for them. The Rand corporation has a list of 50 books for thinking about the future. (http://www.rand.org/pardee/50books/) These offer insights into the past and present and into the minds of successful futurists.
The one thing you will notice about successful futurists is that they don't go overboard predicting killer electronic e coli yogurts. Rather, they outline the likely changes in society and provide suggested remedies for foreseeable problems as well as suggested directions for societal growth.
The area of futurism is very interesting and a strong futurist school of thought is vital to our success as a society. Cranks who like to come up with doomsday scenarios do the entire field a disservice.
Re:And flying cars and moonbases (Score:4, Informative)
Some of these trends are predictably reliable, though. Moore's Law is by no means perfect, but it's extremely likely that computers will continue to grow in processing power at a steady, exponential rate, at least for the next few decades.
The problem is that some - including the typically brilliant Ray Kurzweil - believe that AI is limited by computational power. I don't believe that's the case. I believe that AI is limited by a woefully primitive understanding of several components of intelligence. It is impossible to produce artistic, emotive, sentient machines by applying today's AI models to tomorrow's supercomputers.
Reliable predictions:
- Computers will continue to scale up in power.
- AI models will continue to evolve.
- Thanks to (2), We will eventually succeed at modeling the individual components of intelligence.
- Thanks to (1) and (3), we will eventually produce truly intelligent machines.
That's the most any futurologist can tell you about AI. Anyone who promises more is trying to sell you their book.- David Stein
Witch Doctors, Futurologists, and Cranks (Score:5, Interesting)
I predict that in 2015, this guy will still be making predictions. His track record will be no better than random probability would have resolved. The time you have spent reading his predictions and even this response is time out of your life that you will never recover, and reading it will not put you to any better advantage than if you had not.
Re: (Score:3, Informative)
This is the KICKER (Score:3, Funny)
Maybe not, but I'll have the job debugging all the mistakes the androids will have in their code. They'll be outsourcing debug work to us humans.
Re: (Score:3, Interesting)
And the funny thing is that no one realizes how many times it's happened to different degrees.
"You'll just describe
Man's a fool (Score:3, Insightful)
The other side of AI says that "my brain is magic, and I'm really smart and you can't possibly produce a robot as
clever as me". I don't subscribe to that one - I think that's nonsense.
Tells me all I need to know about this guy's predictions.
He fails to understand that in the 40+ year history of AI research noone has demonstrated even the inklings or foundations upon which actual AI can be built upon.
They may be nothing special about the human mind, but what ever the case is, we certainly havent figured it out yet. It's more likely that we'll have cold fusion by 2015 than AI.
Re: (Score:3, Insightful)
How does he misunderstand that? All he's saying is that there isn't any sort of magical power
Re:Yep (Score:5, Insightful)
Re: (Score:3, Interesting)
Lollipop! (Score:3, Interesting)
Uh, I thought that, explaining what you want to a computer, is precissely what programming is all about. Isn't source code a program's best specification? What are programmers doing if not explaining what they want from the computer?
When someone says "I want a programming language in which I need only say what I wish done," give him a lollipop.
Re: (Score:3, Insightful)
Clearly, between this, and the AI prediction, this guy is completely unaware of computing history.
Only a fool would try to predict the future with no knowledge of the past.
Dijkstra dis
"Futurology" is bunk (Score:5, Insightful)
In 1955 Heinlein, in Revolt in 2100, had the protagonist heading to "the Republic of Hawaii", not able to forsee that fpur years later it would become a state.
Roddenberry had automatic doors, cell phones, and flat screen monitors 200 years in the future rather than 30 years later (now). His writers had McCoy give Kirk a pair of reading glasses in Star Trek IV, not forseeing that twenty years later the multifocus IOD would be developed.
This guy says we'll have six hundred million androids in ten years. He doesn't understand computers, or that AI is just simulation. "I'm in the 30-40% camp that believes that there's really not anything magical about the human brain." But he doesn't see that it is analog, and that thoughts, memories, and emotions are chemical reactions while digital computers are complex abacuses working exactly like an abacus (except it ises base 2 instead of base 10).
He talks of that Warwick guy - "Kevin isn't really the first human cyborg". Nope, he isn't. Vice President Cheney is a cyborg, as he has a device in his heart. I'm one, as I have a device in my left eye (the aformentioned IOD). People have artificial hips and knees. "Captain Cyborg" isn't really a real cyborg, he's a moron like the writer of TFA.
Nothing to see here - at least, nothing for anyone intelligent to see here.
Re: (Score:3, Informative)
http://en.wikipedia.org/wiki/Darpa_grand_challenge [wikipedia.org]
There was a competition of self-driving cars (or SUV's, mostly, and one big truck) put on by DARPA last year, and
Self-Programming Computers, HA! (Score:4, Insightful)
Boy have I heard this one before. It just used to be that computer languages would become so simple that the profession of Programmer would disappear because everyone would just be able to write their own programs. Sure hasn't happened yet.
Someone once famously said: Computers are useless, they can only give answers.
The problem here is, even if you had a computer like the one described here, you still need to be able to understand your problem well enough to cogently explain it to your computer. And that's where most people will fail. They don't understand their problems in the first place, and have no idea how to communicate the solutions they actually need.
Re: (Score:3, Insightful)
$7 for a computer - outrageous (Score:5, Insightful)
item# 172008
http://www.officedepot.com/ddSKU.do?level=SK&id=17 2008&x=0&Ntt=organizer&y=0&uniqueSearchFlag=true&A n=text [officedepot.com]
A lot depends on how you define a computer, but think about what this would have been like in 1970?
Smart Yogurt (Score:5, Funny)
Strong AI is total fantasy. (Score:3, Insightful)
The problem is that people who want to build strong AI are trying to do in decades what took nature billions of years. Certainly, directly engineering something is usually faster than evolving it, but even orders of magnitude speed up won't have strong AI systems any time soon.
Since the dawn of computing, people have been assuming that once computers got fast enough, the AI problems would just solve themselves. The problem is that we're talking about very hard problems. Things that are easy for us (walking, visually recognizing objects, etc.) are hard for computers. Things that are hard for us (math, data processing, etc.) are easy for computers. Why? To do things that are hard for us, humans have developed, over thousands of years, detailed and exacting formalisms. Math has axioms, a syntax, and a set of mechanical processes to carry about. Even complicated proofs involve an extraordinary amount of simple symbol-pushing that a computer could do easily. Computers are based on exactly those same formalisms, so it makes sense that it would be easy to program a computer to do those things. Computers are NOT, however, built anything like how the human brain works, and that's why AI researchers use neural nets and genetic algorithms for so many things.
Long before we're plagued by computers thinking for themselves, demanding rights, and taking over the world, we'll simply continue to be plagued more and more by increasingly catastrophic bugs introduced into increasingly more complex applications. Far from having autonomy, our bug-ridden software does exactly what it was coded to do, right or wrong, and we'll suffer from it. And all along the way, the blame for the problems will fall squarely on the human engineers who made the mistakes in the first place.
Re: (Score:3, Insightful)
Re: (Score:3, Insightful)
Re:Robot brains getting Master Degrees in 20 years (Score:5, Insightful)
Like I said... I don't doubt that eventually we'll develop a computer that can match the processing power of the human brain. But I doubt it'll be soon. It *might* be within my lifetime, but I'm not holding my breath on that one... The brain isn't magical, it's a trillion-core symmetric computer with a staggering memory retention and bandwidth, and a programming so complicated that we're nowhere near matching it. Oh, and that's not mentionning that the brain doesn't work in binary switches, either. It works in chemical switches, with about 50 possible states running in parallel.... Some day, we'll beat out the human brain with a computer. Humans are just too arrogant to believe that we can't, and so somebody will eventually do it. But it's not going to be tomorrow.
Re:Um, this "futurologist" is a moron... (Score:4, Insightful)
Frankly, making broad statements like that sound an awful lot like insisting 640kb of RAM ought to be enough for anyone, or that a computer will never be smaller than a gymnasium, or that man will never fly or travel to the moon.
Re: (Score:3, Interesting)
The confusing thing about all of his "yogurt"-preditctions is that they are internally inconsistant. At first he discusses how electrically-active bact
More than just "not in the direction" (Score:5, Insightful)
But solving something that's NP-complete is not "just a little more difficult" than writing a word processor or an OS. It's so much harder that we need a totally new theoretical framework. Faster processors aren't enough to get us there. And the theoretical breakthroughs come a whole lot less frequently than processor speed increases.
Flying cars? You know, we could probably do that today. It's just a personal STOL aircraft, basically. We can solve the technological problems there. What we can't solve is the rest of it. Between the power requirements (cost) and the driver knowledge needed to operate it, the market size is too small to be worth the effort to create such a beast.
AI? We have the computers that could run the code (maybe). We don't know how to write the code. We probably won't know how next decade, either, or the decade after that.
Smart bacteria? We could perhaps create them. Making them spy on keypresses? Possible. Finding the data you want in the stream of data coming from a trillion (or quadrillion) bacteria? He seems not to have addressed that one.
Sending cans to other parts of the solar system? We've done that. Permanent colonies? It's a lot harder than just sending a bigger can with more stuff in it.
We have breakthroughs in one area (CPUs, for instance) and people assume that other, related problems must be "only a little harder" and therefore about to be solved. But problems differ enormously in difficulty; the level of breakthroughs that we have now is nowhere near what we need for certain problems.
Re:The Color of America? (Score:4, Insightful)
You do realize how fluid the definition of "white" is over the decades, right? A century ago, most of Europe itself wasn't considered "white," especially the southern and eastern bits. When you get right down to it, the only reason Italians, Pols, or even the Irish are considered "white" nowadays is because their emigrants have made a name for themselves in "white" countries like the United States.
Immigrant-friendly "white" countries have taken in plenty of disparate newcomers, far removed from the populations they tried to integrate into, and instead of these countries ceasing to be considered "white" (as many xenophobic contemporaries have always feared), the definition of "white" has simply expanded.