Ensign Steve waxes philosophical on the Singularity, a thrad by Ensign Steve

[Ensign Steve]

Post the zeroeth wherein I hedge a whole bunch before post the first

I like this topic a lot. Seriously, I think about it probably 99% of the time. I have wanted to start a thrad and type about it at great length for a while, but I keep not doing so for several reasons:

- It is super indulgent and navel-gazey, and it's hard for me to take myself seriously for very long when I get like this, at least in public. Call it social conditioning or faux modesty, but it is what it is.

- There is no shortage of insight porn on the internet, and I didn't feel like I needed to add to the noise, especially when people with far more effective rhetorical style have said the same or similar things already. George Carlin I am not.

- I fear (probably rationally) that it will be riddled with assumptions and unchecked bias on my part, and I don't want to offend anybody or embarrass myself. What if it turns out I'm not as awesome and enlightened as I think I am?!

- Whenever I try, I start with some sort of thesis to maintain or conclusion to drive toward, but once I start unloading it branches exponentially and I find myself trying to compose Ensign Steve's Grand Unified Theory of Everything. That's when I get overwhelmed and quit.

Okay, so, hedges have been hedged. Standby for insight. This is in Philosophy and not Computers and Technology, for reasons that I hope will be obvious, but it has a whole lot to do with computers, so go ahead and start getting excited about that.

[Crumb]

[ZEZOZE]

i am excited about this.

[ZEZOZE]

wait.

this isn't a revolution is it? i don't like them revolutions. they leave me lost and feeling like i'm going in circles. so if you can keep it going in a straight line i'd appreciate it. i'm on drugs, ya know.

[Ensign Steve]

Post the first a little bit of background

There's a great introduction to what the hell I mean when I say "the singularity" at this link:

IO9: What Is The Singularity And Will You Live To See It?

Basically, the singularity refers to a theoretical point in the future when artificial intelligence created by humans becomes so intelligent that we can no longer control or even comprehend it. The word "singularity" is borrowed from astrophysics, and has to do with the event horizon around a black hole. It's something like how since no light or information can escape from the gravity of a black hole, we are not able to observe, measure, or perceive anything on the other side of the event horizon (IANAPhysicist, btw). Similarly, since the artificial intelligence will be beyond our comprehension, it is impossible for us to predict anything about our future after the singularity.

It is a concept that has been played around with a lot in science fiction, like The Terminator, The Matrix, and Battlestar Galactica, just to name a few. Pretty much any time the machines rise up and enslave humanity, that's the singularity. In Terminator, they gave it a date. Judgement Day was August 29, 1997. In Battlestar Galactica, it has a sort of BC/AD connotation to it, where it marks the end of one era and the beginning of another. I think it does, anyway, I didn't actually watch BSG, but I loved Caprica.

Here's an article that explores the singularity as the rapture of the nerds, which I think is dead on. A lot of us (nerds) don't believe in the Rapture of the Bible, or the Mayans, or anything supernatural. This gives us something natural (as opposed to supernatural, not as opposed to artificial) that we can believe in without compromising our naturalistic worldview. I think there's something profoundly human about apocalyptic thinking.

(mini-hedge: I'm fairly aware that my point of view on this one is not universal, and this is one of those assumptions I was worried about where I project my perspective on everybody and call it "profoundly human" when I'm probably in the minority. After all, it is the most schizoid among us that tend to be the most apocalyptical, from the foil-hatted homeless man to your average Rapture Ready member.)

This is the part where I'm suddenly relieved that I already banged out my whole "death as the great equalizer" diatribe, because that just saved me from one hell of a tangent. But the short version is that it would take a lot of the sting out of my death if I knew that also everybody else was going to die too and not just move on without me. Sick, I know, but the rapture types have been crowing about that shit for millennia.

Okay, I think that's almost enough background. There are other, less apocalyptic versions of the theory as well, like everybody turning themselves into cyborgs or variations on humans and the AI living in harmony or what have you. The entire point of it is that, by definition, we can't predict what will happen. That's what makes it so fun to explore in fiction.

[But]

Originally Posted by Ensign Steve

Post the first a little bit of background

There's a great introduction to what the hell I mean when I say "the singularity" at this link:

IO9: What Is The Singularity And Will You Live To See It?

Basically, the singularity refers to a theoretical point in the future when artificial intelligence created by humans becomes so intelligent that we can no longer control or even comprehend it.

OK. Here are the biplanes I mentioned. They assume that we are the same we as today.

The word "singularity" is borrowed from astrophysics, and has to do with the event horizon around a black hole.

Actually, it's a pole, which is a point where a function tends to infinity. I remember one good point from a talk of Kurzweil, that the exponential growth like we see in Moore's law can be observed in all kinds of other areas of technology too. If you look carefully, you'll notice that exponential functions don't have singularities. The point at the center of a black hole is a singularity, but only because we don't have a working theory of quantum gravity yet. That's where our known laws of physics break down, that's where the analogy breaks down, although it never really took off.

It's something like how since no light or information can escape from the gravity of a black hole, we are not able to observe, measure, or perceive anything on the other side of the event horizon (IANAPhysicist, btw). Similarly, since the artificial intelligence will be beyond our comprehension, it is impossible for us to predict anything about our future after the singularity.

Then they should call it the Event Horizon. It's not like that sounds like anything less mysterious that science fiction nerds can't wank over.

It is a concept that has been played around with a lot in science fiction, like The Terminator, The Matrix, and Battlestar Galactica, just to name a few. Pretty much any time the machines rise up and enslave humanity, that's the singularity. In Terminator, they gave it a date. Judgement Day was August 29, 1997.

In the Terminator, they have a nice twist. That is something (maybe) closer to a singularity, but still nothing like one, namely closed timelike curves. Having one of those enables you to solve NP-complete and even PSPACE-complete problems in polynomial time.

Here's an article that explores the singularity as the rapture of the nerds, which I think is dead on. A lot of us (nerds) don't believe in the Rapture of the Bible, or the Mayans,

OK

or anything supernatural.

Yeah right.

[Iacchus]

Originally Posted by Ensign Steve

It is a concept that has been played around with a lot in science fiction, like The Terminator, The Matrix, and Battlestar Galactica, just to name a few. Pretty much any time the machines rise up and enslave humanity, that's the singularity.

Schlachthof funf!

Slaughterhouse Five Dresden - YouTube

By the way, the Tralfamadorians in this movie are the good guys.

[willi]

Originally Posted by Ensign Steve

Post the first a little bit of background

There's a great introduction to what the hell I mean when I say "the singularity" at this link:

IO9: What Is The Singularity And Will You Live To See It?

Basically, the singularity refers to a theoretical point in the future when artificial intelligence created by humans becomes so intelligent that we can no longer control or even comprehend it.What about evolution?Do you consider it in your theory?Won't our brain evolve with time as well?If you're referring to robots, I think we might reach that state where robots are not controllable when we allow the robots to create, innovate etc..If we simply gave them a set of command that they have to perform then we wouldn't be in danger. How do you think we can develop robots that have consciousness like our own when we can't even figure out our own brains?Interesting topic ensign

It is a concept that has been played around with a lot in science fiction, like The Terminator, The Matrix, and Battlestar Galactica, just to name a few. Pretty much any time the machines rise up and enslave humanity, that's the singularity. In Terminator, they gave it a date. Judgement Day was August 29, 1997. In Battlestar Galactica, it has a sort of BC/AD connotation to it, where it marks the end of one era and the beginning of another. I think it does, anyway, I didn't actually watch BSG, but I loved Caprica.If you've seen Spielberg's A.I artificial intelligence, it's roughly a similar idea as well. A company that usually produces robots that carry out house functions etc..decide to create a robot kid!And this robot kid can love, and it will be a substitute for people who can't have kids or anything

answers in bold, my first reply to your first post

[Ensign Steve]

Post the second wherein I introduce my thesis, which is either totally mind-blowing, tediously obvious, or something in between

So the angle I've been playing around with the most recently, is that this event is not in the future anymore, that in fact it has already happened.

When I was in high school, my physics teacher tried very hard to impress on us that there was no such thing as The Big One, referring to the giant earthquake that was predicted to destroy California. He said we have "big ones" all the time, and they are devastating, and we recover from them, learn from them, and move on. That's the apocalyptical thinking again. We want things to happen in huge discrete events. The big one, the rapture, the singularity, the fiscal cliff.

When something big happens all at once, we remember it. Black Tuesday, Pearl Harbor, the JFK assassination, the Moon Landing, September 11. But for every Hurricane Katrina story you have, somebody else has an Ivan story or a Sandy story. There have been so many The Big Ones in California that I can't even remember which ones I experienced and which I just read about. Events that individually should actually be very big deals become less significant on repetition. Technological advancement is the same way, from the wheel and agriculture all the way up through the industrial revolution to today.

When Vernor Vinge proposed the term "Technological Singularity" in 1993, he described it like so:

It is a point where our old models must be discarded and a new reality rules. As we move closer to this point, it will loom vaster and vaster over human affairs till the notion becomes a commonplace. Yet when it finally happens it may still be a great surprise and a greater unknown.

Think about what your world was like in 1993. Personally, I didn't own my first computer until 1994. If your 1993 self could see you today, what would you think? Would you even be able to comprehend your job, your social relationships, your priorities, your possessions, or your passions?

This is the slideshow I was clicking through earlier this evening when I decided it was time to finally try to start banging some of this out:

2012 KPCB Internet Trends Year-End Update

Slides 27 - 58 contain a series of side by side "then and now" images about the way we do things differently than just a couple decades ago. I know that some of us have adopted technology into our lives in different degrees than others. Personally, I am fairly mind-blown that I no longer have any use for a dictionary, road maps, or a filing cabinet. I can't even remember the last time I walked into a bank. And kids these days () don't even know what the hell a Dewey Decimal system is.

This insight pornographer coined the phrase "manufactured normalcy field" to describe some about how we can adopt these changes so gradually that stuff that was totally inconceivable to us just just a few years ago is now considered completely normal. There's tons of overlap with skeuomorphism, too, if you really want to go down the rabbit hole.

"But, Ensign Steve," I hear you thinking, "technological innovations happen all the time. Look how much the automobile changed our world. But the singularity is about artificial intelligence. Those apps are great and all, but machines are still dumb as hell. They're certainly not more intelligent than humans!"

You're absolutely right and it was very astute of you to bring it up. Part of me wants to calm down and just figure that this is part of getting older. I grew up never knowing how to use a loom or a butter churn, and I came out just fine. But there's other stuff, specific stuff, that I think makes the current stuff especially singular. I'm going to take a break before I dive into it.

[Ensign Steve]

Post the third wherein I get as technical as I'm going to get (I hope)

I write code at a relatively low level, close to the hardware where I can really get my hands dirty. I feel like I have more fine-grained control of what the machine is doing, physically, not just in the abstract. I'm aware that that feeling is completely illusory, but I have fun with it, and it works for me and my way of thinking.

Compiling, running, and debugging code at that level does not always give me the luxury of error messages, and even when it does, they are what one might charitably describe as "cryptic". When I google those cryptic messages, more often than not the result is a question posted to stack overflow, and the top reply is someone spotlighting some syntax error in the code. The asker replies, "Thanks! That fixed it!" and that's it. Nobody is any wiser about what was going wrong or what the message actually means. They don't care why the fix works, they just care that it works.

I found this attitude frustrating (classic transference: It's debugging low-level code that's frustrating, challenging and rewarding though it may be) and it made me feel pretty snobbish. Why would someone want to program if they don't even care how computers work? I bought into this stereotype of programmers as hackers, tinkerers, puzzlers, and there was no room in my elite little club for higher-level creative types. Designers and abstract thinkers who just want the hardware to get out of their way so they can create. Or, god forbid, maybe some people are just trying to learn a very marketable skill and make a decent living.

-- begin tangent --

And it's not like I even know how computers work. Nobody does. It's like pea commented on the Pinterest (HA! you weren't sure if I was going to call you out or not, were you?):

It's gotten to the point in a lot of things where we don't know how our computers work, though, too. With relatively simple heuristics, you can usually figure it out at least, but when you get into AIs like algorithmic stock trading or even natural language processing and capacity provisioning and other 'squishier' (less actually squishy than they are too complex to be articulated) artificial intelligence applications, we actually don't know how computers are making decisions.

True all that. I attended a meeting the other day where one guy presented this thing about cyberinfrastructure, and he was talking about how the entire thing is a giant pile of hacked together spaghetti code left behind by generations of graduate students building it one piece at a time. It was exactly the same way in the Air Force. In my experience, yeah, that's pretty much how all of it works. Every programmer fantasizes about going back to the beginning and doing it right this time, but that never happens. The temporary solution becomes the solution. I'm talking all up and down the stack, too, from hardware and architecture to operating systems to applications and the communication between them.

Also, it's everybody's job to make their part of the stack transparent to the stack above it. My friend tweeted this thing one time, and I can't find it right now, but it was something like "abstraction is not about being imprecise. it's about being totally precise at every level" ... or something like that. Application developers have the illusion of an infinite, contiguous memory space because someone else implemented virtual memory for them. There's nobody who knows how the whole damn thing works from the bottom to the top.

There's big-time natural selection, too, only totally accelerated, because instead of waiting for random lucky mutations, we get to guide the evolution ourselves. If it works, we keep and build on. If we figure out how to make it better and faster, that becomes the new way. If something is a piece of crap that nobody wants, it will die out. Different architectures become different fitness landscapes. The workings of a mobile phone are totally different that those of a server farm, but they share common distant ancestors. Their genes are in their algorithms.

-- end tangent --

Backing up a little bit to the nineties, we had this world-wide computer network, and humans communicating over that network. It was mostly military at first, I think. Then universities and possibly other government? I'm not even sure. Maybe some hacker nerds trolling usenet. Definitely no women.

That's when I was just getting started in college, and I had my first computer. They gave me a unix login with an email account and a web space. I got it for being in engineering, they weren't even giving them to the English and history majors yet. I didn't know what any of this shit was, but then I got on there and realized there were other people on there and I could talk to them. I figured out how to write shell scripts and download dirty pictures and make web pages, whatever the point of those things were. Here it is 18 years later, and my second-favorite hobby is still typing random shit at strangers until the wee hours of the morning.

So we have this network, which is a great forum for programmers to collaborate and crowdsource their debugging, like on stack overflow. When individuals no longer had to purchase books or take classes or spend hours debugging cryptic error message, it really lowered the barrier to entry. Anyone could fire up a browser, look at tutorials, and experiment with tons of different languages and applications. Best of all, we had 24/7 access to live help in real time, for free. Programming became more collaborative, as well as more accessible to people who tack more to the arty/social side of intelligence than the analytical/mathy types that I associated with programming. Applications were beautiful as well as functional, and had more of a human touch.

Eh, this isn't really a good stopping point, but I'm getting tired and squirrely. I'm pretty sure I'm done with the technical part, at least. If I'm not here when you wake up, talk amongst yourselves and maybe I'll remember tomorrow where the hell I was going with that.

[But]

Hmm, I'm all like, this kind of angle is interesting and everything, but I don't perceive in my refactoring singular way anything more than the usual superficial twaddle by people who like to hear themselves think. Yeah, smoke a doobie and think some deep thoughts geniuses (this is not directed at you ES).

The problem is they are all extrapolating wildly (what else can you do?) and I would bet that they end up completely missing the point again, as you would expect. It's like the question, why haven't any aliens visited us yet? Hmm, maybe there comes a point where civilizations destroy themselves because the gap between how easy it is to build stuff and destroy stuff inevitably gets bigger and bigger. Or maybe there comes a different point where every sufficiently advanced civilization realizes that (for some reason unknown to us now, extra dimensions? multiverse?) traveling to other planets is the stupidest, most pointless thing you could imagine, like banging your head against the wall to see the stars.

Or what is most likely, you don't know what's going to happen because you don't know what's going to make the stuff that's going to happen happen, but it's probably all going to get even more more, like faster and bigger and more extreme. Or maybe not.

It's like when in the past they imagined the future to be full of airplanes, but they would all be biplanes. And they missed the most important things.

[slimshady2357]

Originally Posted by Ensign Steve

Post the zeroeth wherein I hedge a whole bunch before post the first

I like this topic a lot. Seriously, I think about it probably 99% of the time. I have wanted to start a thrad and type about it at great length for a while, but I keep not doing so for several reasons:

- It is super indulgent and navel-gazey, and it's hard for me to take myself seriously for very long when I get like this, at least in public. Call it social conditioning or faux modesty, but it is what it is.

- There is no shortage of insight porn on the internet, and I didn't feel like I needed to add to the noise, especially when people with far more effective rhetorical style have said the same or similar things already. George Carlin I am not.

- I fear (probably rationally) that it will be riddled with assumptions and unchecked bias on my part, and I don't want to offend anybody or embarrass myself. What if it turns out I'm not as awesome and enlightened as I think I am?!

- Whenever I try, I start with some sort of thesis to maintain or conclusion to drive toward, but once I start unloading it branches exponentially and I find myself trying to compose Ensign Steve's Grand Unified Theory of Everything. That's when I get overwhelmed and quit.

I got this far and I was all like "Awesome a thread about the singularity, the beginning of existence (possibly), about what could have caused it, what it all might mean.... and ES starting us off with her thoughts, sounds... awesome!"

And then I read this:

Okay, so, hedges have been hedged. Standby for insight. This is in Philosophy and not Computers and Technology, for reasons that I hope will be obvious, but it has a whole lot to do with computers, so go ahead and start getting excited about that.

And I didn't even blink.

Then I read posts the first, second and third and was just as excited! Something totally different yet equally as interesting. I've never heard of it called the singularity before... I've got some reading to do

[lisarea]

I tend to agree that it's not going to be some headline event. We're not going to wake up and have our newsfeeds announce that the Technological Singularity has arrived or anything; and I do think that, depending on your definition, we're already there.

We already have systems that make decisions with minimal human intervention. That is why the stock market has flash crashes. It's not that the systems are inarticulable or anything. That's I think a pretty common misperception, that if we don't know how something works, that means it's some kind of woo or magic. It's not, though.

It's like human intelligence. Also not woo or magic, but it's too complex to articulate and sort out and explain. You need incredible brain plasticity to take in all the knowledge required to use language optimally, and people cannot even explain what they're doing when they do. And computers work the same way brains do.

So here are two fun freaky facts.

One. Transient global amnesia. This is an apparently common condition, where part of your brain experiences some kind of malfunction, and starts rebooting itself over and over, running a POST process, where the person goes through a script, asking specific questions and running through the same bearings getting information about recent events. The process gets a little longer each time, getting further along in the process, but while it is happening, they always start at the same place.

Here is video of a lady with TGA, rebooting:

It's Tuesday! (Transient Global Amnesia) - YouTube

from this episode of Radio Lab, which listen to it. The doctor is really cool, all planning how if it happens to him, he's going to see if he can break out of the POST process.

Two. The other thing. So you know how people with schizophrenia often have delusions that there are machines controlling or reading their thoughts? Those are called influencing machines, and they way way predate computers. Way before we had anything anyone would really recognize as a thinking machine, people with schizophrenia were imagining thinking machines. And not just randomly or anything, either. The first illustration at that link is of an influencing loom. Loom! Looms are the precursors to the first machines we recognize today as computers. So there's something about the way that machines we design function that we recognize at some level as being like the way we function. Which, of course, makes sense that humangs would design systems that mimic our own systems, whether consciously or not.

This is why this works.

(Which brings us to the problem of privilege and the 'digital divide,' wherein remind me to argue that children who have too easy access to prepared consumer content might be at a fair amount of disadvantage, too, as opposed to kids who have to learn how to make their computers work.)

So other computers, just like us, learn best and most effectively not through prescriptive instruction, but through immersion. And they do. After a certain amount of rudimentary instruction, artificial intelligence systems observe and make associations to mimic and interpret human language and other knowledge.

And what with the internet now being so gigantic and so rich with trivial details, it's a goldmine. There is information out there about even the most banal and abstract topics that is just ready for gleaning. So not only do machines have access to encyclopedic information about specific topics, but they have tweets about what people had for breakfast, Facebook discussions illustrating human communications and relationships, Pinterest boards demonstrating human aesthetics and associations.

So according to a lot of people, seed AIs are the big tipping point, like that's when we can officially announce Technological Singularity! But maybe that's not really so much a big disruptive event. We already have some pretty complex and unarticulated heuristics out there doing stuff as it is, mostly as far as I know in relatively limited domains, and also as far as I know, anyway, they still have off switches if nothing else.

But we already have, and have had for a while, computers that learn, and not at all unlike the way we learn. And for technology to advance much, they need to be able to learn independently and without direct human intervention.

Just a note, too, that we blew past the First Law of Robotics a long, long time ago. I'm talking to you, But.

[But]

Originally Posted by lisarea

Just a note, too, that we blew past the First Law of Robotics a long, long time ago. I'm talking to you, But.

Retrofitting the First Law of Robotics - Transcript - Software Freedom Law Center

There, of course, from the beginning, the assumption was that robots would be humanoid. And as it turns out, they’re not. We do after all live commensally with robots now, we do, just as they expected. But the robots we live with don’t have hands and feet, they don’t carry trays of drinks, and they don’t push the vacuum cleaner. At the edge condition, they are the vacuum cleaner. But most of the time, we’re their hands and feet. We embody them. We carry them around with us. They see everything we see, they hear everything we hear, they’re constantly aware of our location, position, velocity, and intention. They mediate our searches, that is to say they know our plans, they consider our dreams, they understand our lives, they even take our questions — like “how do I send flowers to my girlfriend” — transmit them to a great big database in california, and return us answers offered by the helpful wizard behind the curtain.

Who of course is keeping track. These are our robots, and we have everything we ever expected to have from them, except the first law of robotics. You remember how that went right? Deep in the design of the positronic intelligence that made the robot were the laws that governed the ethical boundary between what could and could not be done with androids. The first law, the first law, the one that everything else had to be deduced from was that no robot may ever injure a human being. Robots must take orders from their human owners, except where those orders involve harming a human being. That was assumed to be the principal out of which at the root, down by the NAND gates of the artificial neurophysiology of robot brains, down there where the simplest idea is, you remember for Descartes, it was “cogito ergo sum”, for the robot it was “no robot must ever harm a human being”. We are living commensally with robots but we have no first law of robotics in them, they hurt human beings everyday. Everywhere.

Those injuries range from the trivial to the fatal, to the cosmic. Of course, they’re helping people to charge you more. That’s trivial, right? They’re letting other people know when you need everything from a hamburger to a sexual interaction to a house mortgage, and of course the people on the other end are the repeat players whose calculations about just how much you need, whatever it is, and just how much you’ll pay for it, are being built by the data mining of all the data about everybody that everybody is collecting through the robots.

[But]

[Ensign Steve]

You know what? We're leaving for Jamaica tomorrow and I was supposed to pack and do laundry, but instead I typed all that. So the point is I'm not going to continue this until after Wednesday or so.

Thanks for reading! I'm having fun.

[lisarea]

THE FUCK, STEVE! WHAT THE HELL? NO, YOU CAN'T GO TO JAMAICA. YOU ARE BUSY TALKING ABOUT THE SINGULARITY.

So anyways, here is my thing with the singularity wherein I agree. We already have AI all over the place, albeit mostly in limited domains (or 'narrow AIs').

We gots the algorithmic stock trading, which we know about.

We have capacity provisioning systems that analyze all sorts of disparate data and make predictions about demands on our infrastructure systems (traffic, networks, electric and sewers, etc.), natural language processing systems using machine learning, knowledge bases like recommendation and search engines, and physical computing systems like the Roomba. Even spam filters use heuristics.

Kevin Slavin: How algorithms shape our world | Video on TED.com

We already have a shitton of AI everywhere all the time already is my point. They're mostly limited purpose, but it would be a simple matter of programming to merge systems and to expand their domains to create more general purpose systems. And as far as the whole thing where they exceed our capacity to understand what they're doing, yes, they already do that too.

Not that it's literally impossible to determine forensically how they make decisions, but it's just impractical to do it every time. But that is already how humans work, anyway. We acquire our native languages through immersion, and we cannot articulate how those work, either. We still discover new grammatical rules all the time.

And that's just comparing an AI with individual humangs. AIs also take advantage of the wisdom of crowds. None of this is stuff that people can't or don't do. It's just faster with a computer.

So I was going somewhere with that, but then I got an extremely annoying phone call, and I forgot what I was going to say next, so I'm just going to post this part and try to remember later. Whatever.

[Ensign Steve]

Originally Posted by lisarea

THE FUCK, STEVE! WHAT THE HELL? NO, YOU CAN'T GO TO JAMAICA. YOU ARE BUSY TALKING ABOUT THE SINGULARITY.

I know, right?!

No, seriously, you don't know how badly I need to disconnect from the matrix and get the fuck out of dodge right now. No email, no phone, no facebook, no homework, no programming, no deadlines. I guess that's why they call it a vacation! Also just to get somewhere where the days aren't so short and I can feel the sun on my skin. I has a SAD.

And that's just comparing an AI with individual humangs. AIs also take advantage of the wisdom of crowds.

That has a lot to do with where I was hoping to eventually land. No one person knowing how the whole thing works, a network of builders crowdsourcing their work, and there emerges from that this global consciousness that is more knowledgable on the whole than any one individual. Like, instead of the AI being this separate thing that we can communicate with face to face, instead of a we're an essential part if it, like termites in a colony, or a collection functional proteins with specialized tasks working together to build and maintain the whole. An earth-sized cyborg brain just humming about in space, learning things about itself...

GAH! I don't have time for this shit right now. I'm trying to pack.

[lisarea]

Originally Posted by Ensign Steve

That has a lot to do with where I was hoping to eventually land. No one person knowing how the whole thing works, a network of builders crowdsourcing their work, and there emerges from that this global consciousness that is more knowledgable on the whole than any one individual. Like, instead of the AI being this separate thing that we can communicate with face to face, instead of a we're an essential part if it, like termites in a colony, or a collection functional proteins with specialized tasks working together to build and maintain the whole. An earth-sized cyborg brain just humming about in space, learning things about itself...

I am putting this here for Wednesday. Don't look until then.

Stanford biologist and computer scientist discover the 'anternet' | School of Engineering

[lisarea]

Oh, OK.

I think I was going to say that, like, yeah, while these developments are normal and explicable, they are still an extremely big deal. So while I disagree that they're literally beyond human comprehension (that's limited by time and resources, not ability), which is the standard tipping point, they do have the power to enhance human potential to an extent we haven't seen before.

We already have a sort of convergence going on, where machines are not just replicating and enhancing human intelligence, but other human performance. People with and without disabilities are increasingly able to do more both mentally and physically, to communicate and move around in ways that they haven't before using external computers and implanted devices. So our bodies are pretty literally changing.

And human cognition is no doubt changing to some degree too, as we have information and computational aids available that let us outsource a lot of the things we used to have to occupy ourselves with, like rote memorization and repetitive computations and things, both at an individual level and as a group.

Some of the notions about the singularity I think are a little bit overstated. Like self-awareness and stuff, or about this idea of uploading ourselves to the cloud or something and becoming immortal. I mean, yes, at some level, we are uploading information that could reasonably be defined as some sort of limited purpose humang, but you cannot upload love!

However, we have definitely gotten to a saturation point. We carry extremely powerful computers with at least limited AI functions in our pockets and all scattered throughout our homes, we have computers operating our vehicles and even our bodies. And that is definitely something to be excited and also worried about. Not because the computers themselves are going to become self aware and rise up against us or anything, but because there's another digital divide up past that first one, where there are people who understand and operate and control and manipulate those intelligent systems, and those who are operated and controlled and manipulated by them; and if we don't have a robust and transparent system of oversight put into place to govern those things, it's not really going to matter whether it's self-aware robots or just people operating regular robots that rise up against us.

And I think the big tipping point that we've pretty much definitely passed is the opt out point. I no longer think it's safe or reasonable for a person to opt out of technology, or even to approach it just as a consumer. I mean, not us. We're old. It's too late. But I mean kids. I don't think it's really a good, viable choice for a kid to not learn about technology anymore, because 1) like language, it's much, much harder to pick up as an adult, after you lose your weird kidlike brain plasticity, and 2) if they aren't at least reasonably technically conversant, they're going to be at a huge disadvantage in a lot of different aspects of their lives. It's not just a career path anymore.

[But]

Originally Posted by lisarea

Oh, OK.

I think I was going to say that, like, yeah, while these developments are normal and explicable, they are still an extremely big deal. So while I disagree that they're literally beyond human comprehension (that's limited by time and resources, not ability), which is the standard tipping point, they do have the power to enhance human potential to an extent we haven't seen before.

In a sense, they are beyond human comprehension (there are so many interpretations of every word). Time and resources are limited, but it's also a societal problem (we are science-fiction nurds, so let's say the neural network of our species super-organism is dysfunctional and because of that the information processing.. OK forget that, you know what I mean).

As a society, we don't seem to have that much of an idea of what we're doing there. Actually, ideas abound, but they don't translate into effective action. We are in the middle of a disaster, we have mountains of good data about it, but the information processing is just fucked up. People still behave a lot like a herd of zombies, so it takes only small pushes in particular directions by the professional perception-manipulators ("PR industry") to control everything.

So we have a case where it couldn't be more obvious and the consequences couldn't really be any worse, and it doesn't work. How then are people going to understand the wider implications of letting shiny little things track you and delegate your thinking to server farms, I mean communicating with your friends and twittering and googling and whatever LOL. It's completely hopeless when it's fun and everybody around you is doing it. So in that sense, it's already beyond human comprehension.

However, we have definitely gotten to a saturation point. We carry extremely powerful computers with at least limited AI functions in our pockets and all scattered throughout our homes, we have computers operating our vehicles and even our bodies. And that is definitely something to be excited and also worried about. Not because the computers themselves are going to become self aware and rise up against us or anything, but because there's another digital divide up past that first one, where there are people who understand and operate and control and manipulate those intelligent systems, and those who are operated and controlled and manipulated by them; and if we don't have a robust and transparent system of oversight put into place to govern those things, it's not really going to matter whether it's self-aware robots or just people operating regular robots that rise up against us.

They don't have to rise up against us, we have to rise up against them. They are already in control. Let's call the locked-down, remotely-controlled devices with their secret programming and malicious features the robots. They are robots; they are not controlled by you, they are controlled by someone else. They are controlled by Skynet, and they spy on you. They spy on you and they hurt you. How is that?

[lisarea]

Originally Posted by Dragar

So my opinion is a bit above my station, but I work with a number machine cognition people (one of whom transferred over from a neuroscience academic career path) and seems to align roughly the same with them.

We're in no danger of a singularity any time soon - our computers and our way of programming computers appears to be fundamentlally different to how biological computers work. And I am fairly convinced that our notions of intelligence hinge on that sort of functionality.

And to make things worse, to draw on Ensign Steve's point that we don't understand how our current computers work - we really don't understand how biological computers work.

Well, then, what is the technological singularity that we're in no danger of? The definition that most use is of 'smarter than human intelligence,' or of systems that function in ways we don't understand.

So what does that mean, anyway? Computers have been faster and more reliable than us at certain types of tasks pretty much forever. It's not that humangs--and I mean humangs in the universal or collective sense, not existential--are incapable of understanding how to do math, but we don't feel like it because math is boring and stupid so we let the boring and stupid computers do it instead. Even if a computer is doing some kind of calculation that we literally don't have the collective time for, like if every person in the world were to collaborate on it, it still doesn't make it unfathomable. It's not doing something totally inconceivable, it's just doing too many conceivable things for us to keep up with sans computers.

For technology to actually do something that is beyond human comprehension, it would have to employ some kind of supernatural force.

So again with language. Humans are capable of using natural human languages innately, but we are not capable of describing them. Early attempts at natural language processing were sort of brute force rule-describing programs, but there is only so far we could go with that because of the sheer volume of rules, and because we haven't articulated most of them yet. We could. There are rules. We don't have time for that shit is all. So modern NLP focuses more on machine learning. That is, rather than trying to write down all the rules that govern language, computers observe language as it's used and make their own observations just like human children do. Computers still can't do this at the level that humans do by a long stretch, but they're already doing things that people haven't fully articulated, similar to the way people do.

Originally Posted by But

In a sense, they are beyond human comprehension (there are so many interpretations of every word). Time and resources are limited, but it's also a societal problem (we are science-fiction nurds,[*] so let's say the neural network of our species super-organism is dysfunctional and because of that the information processing.. OK forget that, you know what I mean).

As a society, we don't seem to have that much of an idea of what we're doing there. Actually, ideas abound, but they don't translate into effective action. We are in the middle of a disaster, we have mountains of good data about it, but the information processing is just fucked up. People still behave a lot like a herd of zombies, so it takes only small pushes in particular directions by the professional perception-manipulators ("PR industry") to control everything.

That's why it's important to distinguish between the universal and existential human. One individual person doesn't have to understand something. If that were the standard, in my case for example, the stock market would literally be some kind of magical fantasy land ruled by warlocks or something, because I dunno. I am capable of understanding it, and back when I was young and needed the money, I had some dark chapters in my life where I had to become a little conversant in it, but I am not dedicating any of my precious long term memory on the stupid stock market.

Other people understand it, though, and hypothetically, I could understand it myself if I were forced to. I just dunwanna because that is part of the human understanding that is only for assholes.

And that extends to other human knowledge too.

And there are certain types of information (not all) that are pretty accurately discovered by the wisdom of crowds, but every discrete person in the crowd could be individually wrong.

Of course, not all types of information are like that, and that crowd wisdom can be very effectively manipulated with shiny stuff, like, ummm, Brave New World,** which is exactly what is happening now with consumer model technology, and we should be very very worried about that.

* For the record, I am not a science fiction nurd. I have not really been into science fiction since I was a kid, so I really only know what people are talking about with stuff like Skynet from the context, and from looking shit up on the internet. Science fiction shortcuts are longcuts for poor old Lisa Pea.

** Check me out, adapting my communications in order to accommodate the interests and proclivities of you young hepcats, making references to various Star Tracks and shit.

[But]

Originally Posted by lisarea

That's why it's important to distinguish between the universal and existential human. One individual person doesn't have to understand something.

Do you mean an individual person doesn't have to understand every single detail to understand something?

If that were the standard, in my case for example, the stock market would literally be some kind of magical fantasy land ruled by warlocks or something,

It isn't a fantasy land ruled by warlocks?

I'd drop the "magical" though. But a warlock has to be magical, right? Let's say a fantasy land ruled by people with dark souls.

* For the record, I am not a science fiction nurd.

Nor am I. I meant we are for the purposes of this thread.

I have not really been into science fiction since I was a kid,

LOL, it's the same in my case. The difference is that you're old, and the Terminator was on when I was a kid.

so I really only know what people are talking about with stuff like Skynet from the context,

That's more than enough. I probably got it wrong anyway.

[lisarea]

Originally Posted by But

Do you mean an individual person doesn't have to understand every single detail to understand something?

Well, when you put it that way, it sounds all dumb and shit, so I will award you three points.

Anyways, I just reread what I poasted before, and it sounds like I'm being snarky or confrontational about that "Well, what is the singularity, then?" part, but I am actually serious. I actually really don't understand what the perception is of what the singularity is.

I have not read the Vernor Vinge thing since [calculating....] 1999, and am only just looking at it again now, in light of intervening events; and I don't see anything jumping out at me that hasn't happened unless you interpret the parts where he's talking about 'awake' systems as being sentient, or maybe have some strict definition of the type of 'interface.' (Maybe I think the literal, physical interface might be the biplane.)

The other main interpretation, which I think is what ES was getting at, was the fundamental change in how we do things. Like a great convergence of disruptive technologies. Which has pretty much definitely happened, or at least is well on its way.

So I really can't think of a real argument that the Singularity isn't here or soon to be, based on my understanding of it. I'm sincerely asking what the pivotal criteria are, because I think I'm missing something.

Nor am I. I meant we are for the purposes of this thread.

Ha, so you mean the collective of 'we.' Individually, you and I are still cool guys, though, and Ensign Steve supplies the nurd.

But she is gone until Wednesday.

WEDNESDAY.

ralflorian.jpg

[Dragar]

Originally Posted by lisarea

Well, then, what is the technological singularity that we're in no danger of? The definition that most use is of 'smarter than human intelligence,' or of systems that function in ways we don't understand.

So what does that mean, anyway? Computers have been faster and more reliable than us at certain types of tasks pretty much forever. It's not that humangs--and I mean humangs in the universal or collective sense, not existential--are incapable of understanding how to do math, but we don't feel like it because math is boring and stupid so we let the boring and stupid computers do it instead. Even if a computer is doing some kind of calculation that we literally don't have the collective time for, like if every person in the world were to collaborate on it, it still doesn't make it unfathomable. It's not doing something totally inconceivable, it's just doing too many conceivable things for us to keep up with sans computers.

For technology to actually do something that is beyond human comprehension, it would have to employ some kind of supernatural force.

So again with language. Humans are capable of using natural human languages innately, but we are not capable of describing them. Early attempts at natural language processing were sort of brute force rule-describing programs, but there is only so far we could go with that because of the sheer volume of rules, and because we haven't articulated most of them yet. We could. There are rules. We don't have time for that shit is all. So modern NLP focuses more on machine learning. That is, rather than trying to write down all the rules that govern language, computers observe language as it's used and make their own observations just like human children do. Computers still can't do this at the level that humans do by a long stretch, but they're already doing things that people haven't fully articulated, similar to the way people do.

When most people refer to the singularity, they talk about the point machines become capable of designing more intelligent machines. Hence the rate at which computers progress diverges, or the timescale of development becomes singular (hence the name - a similar thing does happen with black holes, but not at the event horizon unless you work in silly coordinates).

I don't disagree we could make Chinese-room like constructions if we had the time and inclination, and thare are even some who make the case that this would be intelligence of a sort - perhaps even similar to human intelligence, if sufficiently sped up. But I agree with you that a better approach is the inference/machine learning one.

And yes, computers are getting better at that. And I think parallelism is a bigger paradigm shift than people realise when it comes to doing that sort of thing. But it's hard - really hard. And the successes we have are essentially brute force tecniques of doing what biological systems are clearly doing much, much better jobs of. With processor speed no longer doubling every couple of years (far from it), we can't just add more brute force computational power to solve the problem. I'm excited to see what happens next, but pessimistic it will be anything dramatic in machine intelligence within our lifetime.