The Singularity is not coming

Hi dear reader. We meet again.  

Given that you are tech-savvy, by that point you have almost certainly come across the idea of the Singularity [1] as defended by futurists like Ray Kurzweil and Vernor Vinge. As a reminder, it is the notion that, when we are at last able to compile a smarter-than-human artificial intelligence, this AI will in turn manage to improve its own design, and so on, resulting in an out-of control loop of “intelligence explosion” [2] with unpredictable technological consequences. (singularists go on to predict that after this happens we will merge with machines, live forever, upload our minds into computers, etc).

What’s more, this seemingly far-future revolution would happen within just a few decades (2040 is often mentioned), due to the “exponential” rate of progress of science. That this deadline would arrive just in time to save the proponents of the Singularity from old age is just a weird coincidence that ought to be ignored.

Objection, your honor. As a scientist, I find the claim that scientific progress is exponential to be extremely dubious. If I look at my own field, or at any field that I am vaguely familiar with, I observe roughly linear progress —a rate that has typically been going on since as far back as the field’s foundation. “Exponential progress” claims are usually supported by the most bogus metrics, such as the number of US patents filled per year [3] (essentially a fashion utterly decorrelated from scientific progress). 

And as somebody who does AI research, I find the notion of “intelligence explosion” to make exactly zero sense, for reasons reaching back to the very definition of intelligence. But I am not going to argue about that right now, as isn’t even necessary to invalidate the notion of the Singularity. 

A hypothetical self-improving AI would see its own intelligence stagnate soon enough, rather than explode

I would like to simply argue that scientific progress is in fact linear, and this despite the capitalization of past results into current research (“accelerating returns”), and despite an exponentially increasing population of scientists and engineers working on advancing it (resource explosion). And since I don’t want to argue in the realm of opinion, I am going to propose a simple, convincing mathematical model of the progress of science. Using the same model, I’ll point out that a hypothetical self-improving AI would actually see its own research progress and intelligence stagnate soon enough, rather than explode —unless we provide it with exponentially increasing computing resources, in which case it may do linear progress (or even better, given a fast enough exponential rate of resource increase).

The problem of progress 

The resources that we devote to science have been exploding throughout history. The world population of scientists has been doubling roughly every 50 years from 1500 to 1900 [4], and has kept an exponential rate throughout the 20th century also, by what seems to be a factor of 10 every 50 years [5] (it is however unclear whether that growth is still being sustained now [5], and even if still exponential the rate has evidently gotten slower). Likewise the number of papers and patents has followed exponential growth during the past century, unsurprisingly since average paper count per scientist is unlikely to change significantly. 

Besides available researcher brains, the computing resources available to science have followed exponential growth as well (Moore’s law) in the past 50 years.

Furthermore, each new scientific discovery accelerates the pace of science to some extent, because it gives future researcher a better understanding of the field considered, as well as new tools to work with. That’s “accelerating returns”. You get to stand on the shoulders of giants.  

And yet despite all this, if you look at the actual understanding of the world that we have achieved, if you look at real *impact* rather mere volume, the general impression you get is that scientific progress has really been linear. We didn’t “get” physics 50 times better in 1990 compared to 1940. The 1940-1990 progress of physics might even have been slightly less significant than the 1890-1940 one, actually. 

Even in a new, ultra-fast growing field such as computer science, most of the state of the art algorithms currently around date back from the 60s and 70s. Our computers may be 10^6 faster now than in 1960, but our knowledge is only a few times broader, our algorithm libraries only a few times better. Likewise neuroscience and AI have followed slow, noticeably linear growth since the 50s. 

So how do we explain that huge dichotomy between exponential resource growth and perceived linear progress? What’s happening? 

Well, a widely-noticed empirical fact about science is that within a given field, making an impact is getting exponentially more difficult over time. Just look: the big discoveries in quantum physics had all been reaped by 1930. Most of what could be said in Newtonian physics had been said by Newton himself. Information Theory researchers will never surpass in impact the single 1948 paper that founded their field, no matter how long and fruitful their career. The list could go on. 

In fact, the only way you could hope to really matter in science these days is by founding a new field or subfield —by augmenting a research axiom to the point that it frees new potential big discoveries. Imagination trumps intelligence and hard work every time.  

The reason is simply that founders and first comers reap all of the high-impact low-hanging fruit, and thus after each major discovery, the amount of effort to reach the next one gets multiplied by a certain factor. By the time you join a field where 10,000 people have already spent their career, your impact expectation will be low, and your thesis title seriously obscure. 

Let’s come up with a trivial model and see if increase in discovery difficulty really is *exponential* (or inversely, that at equal difficulty, discovery impact is getting exponentially lower —both are strictly equivalent). If you’re already convinced, skip to the next part! 

Science gets exponentially harder  

Let’s consider a finite set of potential discoveries D within a new scientific field (let’s say 100 discoveries). Each potential finding has an “impact” value following a random uniform distribution between two extrema (let’s say, a random integer between 0 and 100), and an “effort necessary for discovery” also following a random uniform distribution (let’s go again for 0-100). The values I’m mentioning are unimportant and are there to allow you to visualize the problem, dear reader, the only meaningful thing is that I chose random uniform distributions in both cases, which I do because I have no data. Anything else than pure randomness would be too wild a guess. 

Now let’s have a population of researchers that each have an equal number of “effort points” to spend on discoveries over their careers, let’s say 100 points. They come in succession, and each researcher tries to maximize the total impact she can make within the effort she can spend. The sum of the effort factors of the discoveries they make must be smaller or equal to their total effort points. Of course, discoveries can’t be made twice.

Well what have we there, if not an iterated Knapsack problem! Let’s try to run that in Matlab a few times with different random initial distributions, and let’s average the results. Why yes I love Matlab. 

Here’s what I get after averaging 100 trials.

Bam! Exponential decrease of discovery impact of each succeeding researcher. The first researcher to the field makes on average a 801 “impact points” contribution to the field, the second 310, the third 246, etc… 

And it doesn’t really matter how many discoveries are available, one hundred or one million, or how much effort researchers are spending, since these parameters only change the scale of evolution of the graph (as long as both are finite —infinite available discoveries would be a different stories, you’d have progress proportional to effort).  

Conclusion so far:  scientific progress in a given field gets exponentially harder, as widely empirically noticed. Of course paper count per researcher will stay about constant despite this (or patent count, though that one tends to increase by fashion). Mostly, what happens is that papers get increasingly narrow and obscure as the field age, also an empirically obvious trend. 

Essentially that’s why scientific progress is linear within a given field: accelerating returns and exponential growth of scientific resources are merely *compensating* for the increasing difficulty of doing science that matters, the two canceling each other out into linear growth.  

As a side-note: in real life, scientific fields (and thus potential discovery pools) are not set in stone, to the contrary the axioms they are built on regularly evolve, freeing new potential important discoveries for that field (or sprouting new subfields, possibly even entirely new fields, like, say, computer science). It would be enough to invalidate the simple model I’m proposing, and I tackle the objection in the last part of this article.

Now onward with another simple model for a self-improving AI.

The case of the self-improving AI 

Let’s have a hypothetical AI that researches artificial intelligence, and that constantly rewrites its own code to incorporate its finding into its own intelligence. I guess it’s an AI that doesn’t have much regard for its own past identity, or at least that considers its quest to be more important. It would probably feel weird to rewrite your own source code, hey, gives me an idea for an associated reward system. Anyway.

Its first discovery takes ΔT of time, and results in an “amount of knowledge” (or “impact”) K that allows a speedup of the time its takes to make the next discovery (at equal “discovery effort” spent). Next discovery would take ΔT/(ft.K). That is to say thats at equal impact, discovery time would be an exponentially decreasing series of factor ft > 1 (the “intelligence explosion” scenario). 

Also, we model discovery impact (at equal effort spent) as an exponentially decreasing series, as validated by our little model above. Of factor fi > 1. 

So after the N-1th discovery, our AI has made a total impact of:

Σ(1:N-1) K/(fi^n) ~ C for N big enough, and and the Nth discovery is done in ~ ΔT/(C.ft). Of value K/(fi^N). So at that point its progress rate is:

impact/(time to produce impact) = C.K.ft/(ΔT.fi^N)

Conclusion: the rate of progress of the AI converges exponentially toward 0, like that of a frustrated grad student. No intelligence explosion for Raymond. (as a side-note, if you give to the AI exponentially increasing resources (of factor fr), it can manage a linear rate of progress if fr~fi, much like what happens with scientific progress at large). Then again, this is an asymptotic behavior, so if the number of available “big” discoveries is large enough, you could observe a locally exponential progress rate at the start of the curve. On the long term, though, progress rate *will* decreases exponentially.

Seems pretty intuitive actually, if you admit that there is only so much that one can discover about intelligence.  

Paradigm shifts and imagination 

As I pointed out earlier, in real life the paradigms on which scientific fields get founded tend to evolve over time, which means that the set of available discoveries that researchers can “chose from” will tend to expand over time. Paradigm shifts increase the scope of research that can be done. For instance, inventing transistors opened up the entirely new field of what to do with them. Or coming up with the basics of quantum physics. And so on. 

I’d like to argue that these paradigm shifts can be modeled using the exact same model that I used for discovery. Yay for recursive models! We start with a finite pool of possible paradigm shifts, we devote exponentially increasing resources to finding them (“resources” meaning “researchers”, since each researcher tries a little to find out-of-the box ideas), and what happens is that 1) shift volume explodes over time, like papers published did earlier, because they’re a linear function of resources, and 2), the real impact at large of each shift decreases exponentially. Resulting in a linear growth of shift impact… which means in turn that the pool of available potential scientific discoveries increases merely linearly. Here’s a graph to illustrate the idea, the graph of linear scientific progress at a larger scale (we don’t picture individual paradigm shifts, instead we abstract them into shifts of equal cumulated impact, that happen at a linear rate): 

Within each potential progress scope freed by a paradigm shift, progress is linear given exponential resources. Scope expansion are linear as well, given exponential resources. 

So what to make of this? 

• Science advances linearly given exploding resources, and thus its pace will slow down in the future as the resources that we devote to it dwindle. Keep that in mind. 
• Also, a piece of advice: as a researcher, you’ve got to work at the 2nd level of discovery, the paradigm level, because it’s so much more impact-efficient. Don’t devote your energy to discovering tiny new things within the old paradigm you inherited from your thesis advisor, rather discover broad news ideas entirely. Be imaginative. 

Intelligence is just a skill, more precisely a meta-skill that defines your ability to get new skills. But imagination is a fucking superpower. Do not rely solely on your intelligence and hard work to make an impact on this world, or even luck, it’s not going to work. After all the total quantity of intelligence and hard work available around is millionfold what you can provide —you’re just a drop of water in the ocean. Rather use your imagination, the one thing that makes you a beautiful unique snowflake. Intelligence and hard work should be merely at the service of our imagination. Think outside of the box. Break out. Shake the axioms. 

The total quantity of intelligence and hard work available around is millionfold what you can provide —you’re just a drop of water in the ocean

And as a side-note, Ray Kurzweil does not get talked about because of his research, but because he used his wild imagination to come up with far-fetched visions of the future : )

Now of course one would need to go out there and gather invalidating or validating data for the mathematical model I presented. But this is not a paper and I don’t have time for this. Take my model for what it’s worth: as something you’ve read on a random blog. If you’ve got a comment, or want to prove me wrong, be sure to post here or send me an email at francois.chollet@wysp.ws!

Follow @fchollet on Twitter

Comments

[Issa Diao]

This is a great article and covers a lot of ground. I will try to state as quickly as possible why I disagree completely with your assumptions: it isn't individual scientific discoveries that drive progress, it is more a question of reaching thresholds where practical adoption of (not-so-new) technology becomes possible.

An obvious example is the telephone. The underlying technology certainly progressed linearly--no argument there. But once a threshold was reached where it became possible for billions of people to carry phones in their pockets, it made a huge impact on society in a relative instant.

I think these examples are everywhere--from medicine/health to 3D-printing to (hopefully) space exploration. I'm not an AI expert, and I hope I'm not being naively optimistic...but it certainly seems like technology is progressing towards a threshold.

[webreac]

Human progress may be linear, but Moore law is exponential since many years. If you have an AI that builds others AI, exponential growth is expected, at least at the beginning. After this phase of exponential growth, the world may have changed substantially.

[Peter]

I agree with you that humanity is unable to produce results at the rate at which we'd like to see. The idea that increasing your intelligence by means of coding is a far-fetched idea and I cannot see a way to write a program that is able to rewrite itself with the goal of making itself smarter. If we reach near human intelligence levels, then hardware will be the factor that makes it "smarter". The ability to recall and store working information is one factor of intelligence that will benefit.

I disagree with regards to intelligence formulae. A loose definition of intelligence is (in my opinion) is the ability to take the knowledge that you have and apply it to problems. You then need to apply hard work and creativity to test your theories and extend your field of knowledge. This is where computers are far superior. If we have a machine with knowledge of all scientific fields, it will be able to brute force itself through many combinations of models. Cross-pollination of scientific fields is (once again in my opinion) where we will see this intelligence explosion.

[Harry Pachty]

When we talk about complex systems and artificial intelligence, we use the word 'emergence' to describe something that comes into existence without being designed in the first place. I do not find this word anywhere in your posting, so anyone who thinks he can tell me something new without having a grasp of the basic principles is close to a shaman chopping the head of a chicken and telling me my fate.

You answered the wrong question ..

[Cognitive Social Web]

I sure wish "emergence" meant something, then perhaps I could use it in my research, because all that emergence stuff sounds seriously powerful!

On a more serious note, I tend to dislike buzzwords.

[Loup Vaillant]

@Harry: beware the word "emergence", it is often a magic word used do avoid explaining something you don't understand. Yep, there could be real emergence of interesting things that weren't designed in the first place, but if you hope to "emerge" something purposefully, you'd better have an idea *how* to emerge it.

Regarding the intelligence flatline, I have some doubts. I get that the effort needed to make the next discovery will be exponentially harder. But why the impacts of such discovery would not compound themselves such that total impact would be the *product* of all impact, and therefore exponential as well? In that case we have 2 exponentials fighting each other, which makes early flatline much less certain than in your model.

This may not apply to current science research, but that's because we humans have constant intelligence (and imagination and all that). A self improving AI is also allowed to increase its own intelligence and imagination, for this is precisely what its discoveries are supposed to be about. (By the way, you seem to separate intelligence and imagination. We're not magic, so there's no reason our imagination cannot be imbued in a machine —though currently, we certainly don't know how to do that).

[Adi]

"Intelligence is a skill, but imagination is a fucking superpower!"
I like that! :)
I would also say imagination is a key to connecting with vast amounts of knowledge and understanding that intelligence can even begin to comprehend.

[Alex]

I've seen Kurzweil graphs, and ones that purport to argue that science hasn't discovered anything since whenever, and the conclusion I got was that the difference between them was simply an a priori decision as to what counted as important science. If you're a computer scientist or a biologist you think we're in a golden age; if you're a physicist, rather less, unless you're in materials or working at CERN.

More interestingly, what do you make of Stanovic & West's notion of intelligence and rationality as independent vectors? An AI would, by definition, be all intelligence and no rationality, a contextless brain in a jar. S&W (and Daniel Kahneman) would argue that was useless.

[PK]

Why do you think we are pumping in exponentially more resources for a linearly rise marginal utility from science? Is it that we value Science more or are funding agencies behind the curve? ;)

[j03]

First you complain of "arbitrary metrics" used by Singularity proponents and then you substitute them with your own anecdata ("If I look at my own field, or at any field that I am vaguely familiar with, I observe roughly linear progress") and arbitrary timelines, ("The 1940-1990 progress of physics might even have been slightly less significant than the 1890-1940 one, actually.")

1990 was over 20 years ago! And less significant by what measure? I thought you were going to produce some scientific proof but I can't even get past the first few paragraphs without rolling my eyes!

[Seumas Mackinnon]

You appear to make three main arguments in your essay:

A) I don't like Kurzweil therefore he's wrong
B) Kurzweil is looking to benefit from the Singularity therefore he's wrong
C) Increased complexity makes it more difficult to reach even more complexity

I don't particularly care about Kurzweil either way, so I'm afraid arguements A & B are wasted on me. They do have the quality of being the most common arguments I've seen, I suppose.

C is more interesting. You make a good case. It does have the minor flaw of being directly contrary to everything we know about actual evolution of intelligence on Earth, but at least you could consider *why* it doesn't work there...

[Joe]

There is only one problem.

Moore's Law completely proves you wrong.

If your argument was valid, then Moore's law wouldn't work, as reaching more complexity is supposed to take longer and longer for the same relative gains.

[Dave Sutter]

I don't think the graph of scientific impact you made is exponential. You found an exponential fit, of course, because you fit an exponential to it. The "experimental" curve in fact falls off faster than an exponential. And it will go to zero. This is because for one thing there is a finite amount to be discovered in your model. So progress slows down when we run out of things to discover. More importantly, it models the difficulty of discovery as an absolute number independent of time. The reason we have an exponential increase in knowledge is that new discoveries become easier based on previous discoveries. In other words, the rate of change of the number of discoveries depends on the number of discoveries that have been made.

[gwern]

> That this deadline would arrive just in time to save the proponents of the Singularity from old age is just a weird coincidence that ought to be ignored.

This is false and unfair of you; it *may* be true of Kurzweil, even though he's 64 with an expected longevity perhaps in the 80s which gives him another 18 years rather than the 28 necessary to reach 2040, but I will assume that we are handwaving this away by beliving that Kurzweil's insane theories about supplementation make him think that he can reach 2040.

But it's not true of people in general.

It's certainly not true of one nest of transhumanist vermin, LessWrong: http://lesswrong.com/lw/8p4/2011_survey_results/5dny
The related SIAI looked for such a horizon effect and Shulman says nothing obvious popped up: http://lesswrong.com/lw/1hn/call_for_new_siai_visiting_fellows_on_a_rolling/1bhc
The FHI survey on the timing is even more clearly not exhibiting this effect as most place the Singularity well after they are dead: http://www.fhi.ox.ac.uk/__data/assets/pdf_file/0015/21516/MI_survey.pdf

Perhaps someone might want to bring up http://www.kk.org/thetechnium/archives/2007/03/the_maesgarreau.php as evidence?

> You will not be surprised to find that in half of the cases, particularly those within the last 50 years, the Singularity is expected to happen before they die – assuming they live to be 100.

The errors in the small table aside ('Hansen'? And Eliezer didn't predict 2005, he said 2005-2020 as a goal and not a prediction), you don't even get to 50% during their lifetime without arbitrarily tacking on 20+ years to life expectancy! So the very examples he's using do not satisfy his 'Maes-Garreau Law'... And why would 50% be evidence of such a bias in the first place, anyway? This is pretty sloppy of Kelly.

(Oh yes, and I should note the irony of this criticism in the first place: people who were predicting it after their lifetime get accused of making their beliefs unfalsifiable! Damned if you do, damned if you don't. One wonders what bias or law would be made up to explain eg. Shane Legge's ~2020 prediction.)

Stay classy, François...

[DMcCunney]

People pointing at Moore's Law as a falsification of this premise need to go back and look at what Moore's Law *is*.

Moore's law is the observation that over the history of computing hardware, the number of transistors on integrated circuits doubles approximately every two years.

We are starting to run into fundamental physical constraints that are making keeping up this rate of progress unlikely, but whether we manage to keep doing it of not, it has nothing to do with above premise and does not negate it.

[Matthew J Price]

Thanks for putting thoughtful work into such an important topic, Cognitive Social Web, though I disagree about a few points and agree about something perhaps more important.

Others have pointed out that your first arguments aren't relevant, and I don't think you meant them to be, so for me, they could have been left out. Yes Kurzweil (and I and everyone) has every incentive to believe that we will be able to live with our families forever in unimaginable luxury and breadth of experience. Bias Granted. However, your main points are more interesting.

You argue that it is due to exponentially increasing resources that we've been able to "compensate" for the increasing difficulty in doing science as a field develops. My first response is: Ok, that's fine. Our resources are going to continue exponentially growing in a few ways:

First, the rest of the world is coming online. Recent forecasts predict 5 billion people will be online in 2020. I would be very surprised if all 8 billion of us weren't on the internet by 2030, but the point is there's going to be a whole world with access to the newest ideas. After everyone who wishes to do science is doing it, then we'll begin to see this drop off and one part of our exponentially increasing resources will be out of the equation.

The next contributor to our exponentially increasing resources are our tools. These, unlike human beings, show no sign of ever halting their progress. We don't need AGI (though it would certainly help) in order to continue expanding the scope of human inquiry as work in various scientific fields becomes automated with both robots and algorithms. You rightly point out that 1 person shouldn't expect to achieve much with their effort and ideas, which are usually expendable. However, that doesn't mean they are producing less than they would have in an earlier era of relative ignorance (and thus a broader domain of inquiry). I agree most fields aren't using tools that are on directly exponential paths (yet); Genomics is a favorite of futurists like Kurzweil because it can benefit so directly from Moore's Law and thus every dollar they throw at their problems will sequence ever more genes as the years go on. We don't need every field to do this, because of my third point

Technologies converge and create new domains of discovery in the "adjacent possible". work in materials produces advances in astronomy, computing, energy, automotive, etc, while achievements in synthetic biology make new work possible in medicine, energy, agriculture, etc. As noted above, the cross pollination could certainly be brute forced by an AI, but even lacking one it will dramatically increase as a function of information transparency, better search and sorting algorithms and hyper specific specialization. To use the metaphor of low hanging fruit, as branches become heavier, they push still other branches lower for our reach. Alternatively, we're building ladders out of the branches we've already picked or growing new orchards, take your allegorical pick.

I think I can accept your thesis, that scientific endeavor is a linear endeavor, but once you allow that it will scale as a function of its inputs, then you've admitted the singularity is a distinct possibility; All we require is that there are uncountably infinite number of scientific propositions, and that our inputs continue their exponential rise atleast for another 30 years, which is, I suppose where we will ultimately differ.

Thanks again, I hope I offered something helpful.

[Shaw]

Interesting argument but not accurate. Even if scientific discoveries grow linearly technological advancements can be exponential. No new science was needed to make the Moore's law work. Similarly, no new science was needed for the Web to grow exponentially. Technology depends on science but it is not the same thing.

[Mindbleach]

Technological progress is not the same as scientific progress. It's not 'the scientific singularity.'

If you perfectly understood consciousness, could you build a digital brain with a billion neurons? I suspect not. Twenty years ago, definitely not. Twenty years from now, why not? That gradient from impossible to trivial is 1% science and 99% engineering.

Sharp rock makes sharper rock. When minds become technology, sharp mind makes sharper mind.

[Mentifex (Arthur T. Murray)]

"Non uno die facta est Roma," OK, I can agree that Rome was not built in one day, but every day I am working to bring about The Technological Singularity yet in this current year of Anno Domini 2012. How? you may splutter. By releasing open-source free-of-charge artificial intelligence as JavaScript http://www.scn.org/~mentifex/AiMind.html in English; as JavaScript http://www.scn.org/~mentifex/Dushka.html in Russian; and as Win32Forth http://www.scn.org/~mentifex/mindforth.txt in English for embodiment in robots. Currently the MindForth AI is the most advanced AI and it has just yesterday on Fri.10.AUG.2012 become capable of, and demonstrative of, self-referential thought -- you can talk to it across the keyboard about itself, then query it for what it knows and remembers about itself (or about you, for that matter :-) so please give The Singularity at least until the end of 2012 to manifest-destiny itself. Thank you. Mentifex (Arthur)

[Cacatua Galerita @Sulphurcocky]

Wow, Most interesting, thanks for your good points. Some things came to mind as I read your article, A singular AI most likely would not specialise, if it is self adaptive? and that you also describe the human condition of development, eventually we humans don't gain enough to continue to develop, we stop until pressured to adapt to different surroundings or experiences.

[David Barbour]

I feel you missed an accounting of how the individual human progresses, and its impact on scientific progress. In particular, the more we know about a subject, the more each human needs to learn before making a significant contribution. An AI does not face this issue of needing to systematically relearn and reteach everything in each generation. It can also make connections much wider (multi-disciplinary) than any individual human.

I do not know whether those will be enough counteract eventual stagnation, but they could certainly create a much longer tail for linear progress.

Also, there is no reason "imagination" cannot itself be automated, e.g. in the form of probabilistic searches or generative grammars. There isn't anything super powerful about imagination.

[Loup Vaillant]

You probably mean "nothing magic", not "nothing super powerful". Being able to implement imagination in a machine doesn't diminish its power. And it looks like our imagination is a quite important part of our optimizing power. We often correlate impact and probability negatively, but I wouldn't in this case.

(Now, imagination may not be that powerful either. But that's a separate question. I'm just saying you don't need imagination to be weak to make your point.)

[David Barbour]

"Being able to implement imagination in a machine doesn't diminish its power."

But it does allow us to quantify its power, and relate it to other intelligence models. To the extent that imagination is "super powerful", I think we could say the same of any other aspect of intelligence.

[Joey1058]

I make no claim to be any kind of scientist or researcher whatsoever. So my comments here are of the wowwee, gee whiz! variety. First off, a refreshing perspective. I like when there is someone to take a weighty theme and just shoot holes in it. I read in the comments twice referencing Moore's law being exponential. Someone will correct me if I'm wrong, but didn't we actually hit Moore's ceiling? Science is cheating. All we're doing now is taking that physical limit, and purposely doubling that final result. To me that is linear. Combining two or more core processors onto one platform. All it's doing is sharing the workload to make the illusion of faster processing speeds. A "don't look behind the curtain" approach.

I've always been of the opinion that we will not experience a "true" singularity. Like Relativity, your speed is in relation to the observer. You might be going light speed, but you notice you're standing still. I believe we're somewhere in the knee of the curve now. But we're not seeing it. Likewise, when it happens, we're not going to see that either. Someday, we're going to reach a point in humanity's existence and science will take a moment to look behind itself and say, "Well, shit, when did I miss my turnoff?".

[Vinney]

"I sure wish "emergence" meant something, then perhaps I could use it in my research, because all that emergence stuff sounds seriously powerful!

On a more serious note, I tend to dislike buzzwords."

Excellent. Perhaps someone is a fan of LessWrong?