Superintellligence

This talk, by Maciej Cegłowski, which reached me through a number of links, is one of the most interesting things I've seen on the web in a long time. I warn you that it's long, but it's not boring. The topic is superintelligence, the idea that computer-based intelligence is growing so fast that it will soon be dominant in our society. The proponents of this idea think this is a wonderful thing. Our author is not so sure.

Here are some short excerpts:

AI Cosplay
The most harmful social effect of AI anxiety is something I call AI cosplay. People who are genuinely persuaded that AI is real and imminent begin behaving like their fantasy of what a hyperintelligent AI would do.
In his book, Bostrom lists six things an AI would have to master to take over the world:

• Intelligence Amplification

• Strategizing

• Social manipulation

• Hacking

• Technology research

• Economic productivity

If you look at AI believers in Silicon Valley, this is the quasi-sociopathic checklist they themselves seem to be working from.
Sam Altman, the man who runs YCombinator, is my favorite example of this archetype. He seems entranced by the idea of reinventing the world from scratch, maximizing impact and personal productivity. He has assigned teams to work on reinventing cities, and is doing secret behind-the-scenes political work to swing the election.
Such skull-and-dagger behavior by the tech elite is going to provoke a backlash by non-technical people who don't like to be manipulated. You can't tug on the levers of power indefinitely before it starts to annoy other people in your democratic society.
I've even seen people in the so-called rationalist community refer to people who they don't think are effective as ‘Non Player Characters’, or NPCs, a term borrowed from video games. This is a horrible way to look at the world.
So I work in an industry where the self-professed rationalists are the craziest ones of all. It's getting me down.,

Incentivizing Crazy This whole field of "study" incentivizes crazy. One of the hallmarks of deep thinking in AI risk is that the more outlandish your ideas, the more credibility it gives you among other enthusiasts. It shows that you have the courage to follow these trains of thought all the way to the last station. Ray Kurzweil, who believes he will never die, has been a Google employee for several years now and is presumably working on that problem. There are a lot of people in Silicon Valley working on truly crazy projects under the cover of money.

Religion 2.0
What it really is is a form of religion. People have called a belief in a technological Singularity the "nerd Apocalypse", and it's true.

It's a clever hack, because instead of believing in God at the outset, you imagine yourself building an entity that is functionally identical with God. This way even committed atheists can rationalize their way into the comforts of faith.
The AI has all the attributes of God: it's omnipotent, omniscient, and either benevolent (if you did your array bounds-checking right), or it is the Devil and you are at its mercy.

Like in any religion, there's even a feeling of urgency. You have to act now! The fate of the world is in the balance!

And of course, they need money!

Because these arguments appeal to religious instincts, once they take hold they are hard to uproot.

Kevin Kelly says: "Nobody ever suggested that Picasso should spend fewer hours painting per picture in order to boost his wealth or improve the economy."

Want to bet?

Despite the snark, I think this is one of the most interesting articles I've read on the web this year or maybe even last year! (smile)  Some years back I came to the conclusion that the general availability of safe clean water was the measure of civilization. I am not backing off that statement, but Kevin Kelly has made me rethink myself. You should read his article, because it is a great push back on the philistinism of our current age. Potential Picassos are being told to get a job all the time, and the cultural infrastructure built by past Picassos and their fans is being dismantled in the name of both morality and prudence, but mainly productivity.

Take a look at these farm houses which I saw under construction in remote areas of Yunnan province China. They were not unusual; farmsteads this size were everywhere in rural China. Note the scale of these massive buildings. Each support post is cut from a single huge tree. The massive earth walls are three stories high and taper toward the top. They are homes for a single extended family built in the traditional Tibetan farmhouse style. They are larger than most middle-class American homes. The extensive wood carvings inside and outside will be painted in garish colors, like this family room shown in a finished home. This area of Yunnan is consider one of the poorer areas in China, and the standard of living of the inhabitants here would be classified as "poor."
Part of the reason is that these homes have no running water, no grid electricity, and no toilets. They don't even have outhouses.
But the farmers and their children who live in these homes all have cell phones, and they have accounts on the Chinese versions of Twitter and Facebook, and recharge via solar panels.
This is important because a recent thought-provoking article by a renowned economist argues that the US economy has not been growing during the internet boom and probably will not grow any more than it has already because computers and the internet are not as productive as the last two industrial revolutions.


You can read the article here: Is U.S. Economic Growth Over? (PDF) by Robert Gordon.
Gordon answers his own question with: Yes, US economic growth is over for a while. I think Robert Gordon is wrong about his conclusion, but I wanted to start with one of the bits of evidence he offers for his view. He is trying to argue that the consequences of the 2nd Industrial Revolution, which bought to common people electricity and plumbing, was far more important than the computers and internet which the 3rd Industrial Revolution has brought us. (Gordon's 1st Industrial revolution was steam and railroads.) As evidence of this claim he offers this hypothetical choice between option A and option B.

With option A you are allowed to keep 2002 electronic technology, including your Windows 98 laptop accessing Amazon, and you can keep running water and indoor toilets; but you can’t use anything invented since 2002. Option B is that you get everything invented in the past decade right up to Facebook, Twitter, and the iPad, but you have to give up running water and indoor toilets. You have to haul the water into your dwelling and carry out the waste. Even at 3am on a rainy night, your only toilet option is a wet and perhaps muddy walk to the outhouse. Which option do you choose?

Gordon then goes on to say:

I have posed this imaginary choice to several audiences in speeches, and the usual reaction is a guffaw, a chuckle, because the preference for Option A is so obvious.

But as I just recounted, Option A is not obvious at all.
The farmers in rural China have chosen cell phones and twitter over toilets and running water. To them, this is not a hypothetical choice at all, but a real one. and they have made their decision in massive numbers. Tens of millions, maybe hundreds of millions, if not billions of people in the rest of Asia, Africa and South America have chosen Option B. You can go to almost any African village to see this. And it is not because they are too poor to afford a toilet. As you can see from these farmers' homes in Yunnan, they definitely could have at least built an outhouse if they found it valuable. (I know they don't have a toilet because I've stayed in many of their homes.) But instead they found the intangible benefits of connection to be greater than the physical comforts of running water.
Most of the poor of the world don't have such access to resources as these Yunnan farmers, but even in their poorer environment they still choose to use their meager cash to purchase the benefits of the 3rd revolution over the benefits of the 2nd revolution. Connection before plumbing. It is an almost universal choice.
This choice may seem difficult for someone who has little experience in the developing world, but in the places were most of the world lives we can plainly see that the fruits of the 3rd generation of automation are at least as, and perhaps more, valuable than some fruits of the 2nd wave of industrialization.
So if people value the benefits of computers and internet so much why don't we see this value reflected in the growth of the US economy? According to Gordon growth has stalled in the internet age. This question was first asked by Robert Solow in 1987 and Gordon's answer is that there are 6 "headwinds," six negative, or contrary forces which deduct growth from the growth due to technology in the US (Gordon reiterates he is only speaking of he US). The six "headwinds" slowing down growth are the aging of the US population, stagnant levels of education, rising inequality, outsourcing and globalization, environmental constraints, and household and government debt. I agree with Gordon about these headwinds, particularly the first one, which he also sees as the most important.
Where Gordon is wrong is his misunderstanding and underestimating of the power of technological growth before it meets these headwinds.
... the 3rd Industrial Revolution is not really computers and the internet, it is the networking of everything.
And in that regime we are just at the beginning of the beginning. We have only begun to connect everything to everything and to make little network minds everywhere. It may take another 80 years for the full affect of this revolution to be revealed.
In the year 2095 when economic grad students are asked to review this paper of Robert Gordon and write about why he was wrong back in 2012, they will say things like "Gordon missed the impact from the real inventions of this revolution: big data, ubiquitous mobile, quantified self, cheap AI, and personal work robots. All of these were far more consequential than stand alone computation, and yet all of them were embryonic and visible when he wrote his paper. He was looking backwards instead of forward."
Finally, Gordon is focused, as most economists, on GDP which measures the amount of "labor saving" that has been accomplished. The more labor you save while making or serving something, the more productive you are. In the calculus of traditional economics productivity equals wealth. Gordon rightly points out that so far the internet has not saved a lot of labor. As I argue in my robot piece in Wired, Better Than Human (not my title), I think the real wealth in the future does not come from saving labor but in creating new kinds of things to do. In this sense long-term wealth depends on making new labor.
Civilization is not just about saving labor but also about "wasting" labor to make art, to make beautiful things, to "waste" time playing, like sports. Nobody ever suggested that Picasso should spend fewer hours painting per picture in order to boost his wealth or improve the economy. The value he added to the economy could not be optimized for productivity. It's hard to shoehorn some of the most important things we do in life into the category of "being productive." Generally any task that can be measured by the metrics of productivity -- output per hour -- is a task we want automation to do. In short, productivity is for robots. Humans excel at wasting time, experimenting, playing, creating, and exploring. None of these fare well under the scrutiny of productivity. That is why science and art are so hard to fund. But they are also the foundation of long-term growth. Yet our notions of jobs, of work, of the economy don't include a lot of space for wasting time, experimenting, playing, creating, and exploring.

 Thanks to Brad DeLong for the heads-up. And Will McLean for this comment.

IT heroes save the day at Nipissing University

Call it involuntary distance education.  But it all worked out for the best.

This morning I woke to the biggest heaviest snowfall I have seen in a long time.  I have a very long country driveway that can be cleared by hand if the time is available, or plowed if our neighbor is available.   Looking out the window I knew that handscraping was not going to be possible. Certainly not before my 10 o'clock class.  My three-hour seminar starting at 12:30? That would depend upon my neighbor.

By 11 o'clock it was pretty clear that he was not going to show up in time.  (It turns out that he was having tractor trouble.) What to do? Cancel another class? That thought really bugged me. This was the last class before our in-class midterm, and there were things I wanted to get done.

One faint possibility occurred to me. What if I used Skype to communicate with the students? It might not be an ideal way to run  a seminar discussion, but surely it would be better than an unplanned disruption. I called the IT helpdesk. Could they set up a computer in the classroom to be a Skype station? Would they?

No hesitation. I am sure they had plenty of other things to do, but with great willingness and competence they set up my classroom so that I could discuss the upcoming term test and run a discussion on queens and other wives in sixth-century France.  I would not have you think it was perfect, but it was better than the alternative: sitting in front of the fire with a glass of wine… Wait a minute, was that a possibility?

I was glad in any case not to lose the class meeting. Thanks, Marg!  Thanks, Greg!  Thanks, whoever else pitched in (that includes at least one student)!

Sometimes this really is a great place to work.

Image:  hmm...

Twenty years late -- The Difference Engine

Twenty years ago, William Gibson and Bruce Sterling published The Difference Engine, an alternate world novel in which actual plans for a large, mechanical computer were actually realized, so that the steam-driven actual Industrial Revolution was hyperpowered by a cybernetic component.  This book was one of the most important influences behind the steampunk "fad," if something so long-lived can be called a mere fad.


Earlier this year I mentioned to some friends that I had not only never read the book, but never even seen a copy.  Soon after that, they gave me one.

I enjoyed the book, when I finally made time to read it, but I could see why several people had told me over the years that it was not a very successful novel (but perhaps it wasn't really meant to be a novel?).  It is certainly no Islands in the Web (a Sterling novel that was one of the best books I read in the 1980s).  But no historian who was raised on science fiction could not like a portrait of Britain where Lord Byron is PM and head of the Radical Industrial Party.

Twenty years late, hats off!

Image:  A difference engine built in the 21st century, at last.

Charles Babbage's Difference Engine

I am fascinated by early 19th-century technological breakthroughs, so I was really pleased to be referred to this NPR piece on the Babbage Difference Engine, which includes audio, text, photo slideshow, and video.

Here's an excerpt:

Charles Babbage, the man whom many consider to be the father of modern computing, never got to complete any of his life's work. The Victorian gentleman was a brilliant mathematician, but he wasn't very good at politics and fundraising, so he never got the financial backing to finish any of his elaborate machine designs. For decades, even his fans weren't certain whether his computing machines would have worked.

But Doron Swade, a former curator at the Science Museum in London, has proven that Babbage wasn't just an eccentric dreamer. Using nothing but materials that would have been available to Babbage in the 1840s, Swade and a group of engineers successfully built Babbage's Difference Engine — and a version is now on display at the Computer History Museum in Mountain View, Calif.

The Difference Engine fills half a gallery and stands taller than most men. It's 5 tons of cast iron, steel and bronze woven together from 8,000 distinct parts. Though it looks like it could be a sculpture, the machine is essentially a giant calculator. Tim Robinson, a docent at the museum, says it's "the first automatic calculating machine."

This engine — made from 162-year-old designs — doesn't have a power pack; it has a hand crank. Robinson works up a sweat as he turns it. "As long as you keep turning that crank, it will produce entirely new results," he says.

Most importantly, the machine produces accurate results. In Babbage's time, England reigned over a vast global empire. To navigate the seas, captains used books filled with calculations — but these equations were all done by fallible human minds.

"If the tables had an error," Robinson says, "a ship could either get lost or run aground, so lives and property were thought to be at stake."

The story goes that Babbage was inspired to create the Difference Engine one day when he came across multiple errors in a book of astronomical calculations. "I wish to God these calculations had been executed by steam!" he exclaimed.

Of course, this episode inspired one of the first steampunk novels.

P.s. It was designed to have a printer! And has one now!

Image: Wikipedia picture of the American exemplar.