Accelerating Future

And this is the Rice dwarf virus of the family Reoviridae. The Reoviridae family is a family of viruses that attack the gastrointestinal system. What you see is a computer model built using X-ray crystallography techniques to image the virus atom-by-atom.

What is fascinating is their similarity in appearance. Parallels between objects at such vastly different scales points to the fact that we live in a low-entropy, fractal universe. This allows us to make inferences about new objects using metaphor and metonymy, and actually have a chance that those inferences are correct. If our universe consisted of radically diverse objects, past experience would be far less useful in predicting the future - in fact it would have been difficult for any form of intelligence to emerge.

Contemporary physics theories assert that our observable universe is only a small part of a much larger region called the multiverse. The multiverse is infinite for all practical purposes, and contains universes of every size and with every conceivable set of natural laws, including those with greater or fewer spacetime dimensions.

Most of these universes lack the conditions necessary to foster life - for example, if the strong force in the atomic nucleus were slightly weaker than it is here, stable atoms would be impossible, and all the matter in the universe would be a diffuse fog. But finding ourselves in a universe fine-tuned to support life should be no surprise. After all, how could our species have popped up anywhere else?

Specific complexity requires order. Our universe has orderly natural laws that allow complexity to accrete without the interference of chaos. It even looks like the product of an evolutionary selection process that favors the existence of life.

The multiverse hypothesis comes from quantum mechanics. In quantum mechanics, everything is a chaotic foam at the lowest level, involving random tunneling effects and principles of uncertainty.

If every universe in the multiverse is part of a huge, randomized quantum manifold (as our evidence suggests), then we should expect disorderly universes to be more common than orderly universes. This “universal majority” is far too chaotic and disorganized to harbor intelligent observers. In the same way that SETI scientists specify a planetary “habitable zone” at a certain distance away from a central star, there is a multiversal “habitable zone” where all conscious beings live.

Think of a cellular automata grid like the Game of Life. The simple, disorganized structures are the first to emerge. For the longest time these are the only structures visible. Then, simple, progressively more organized structures can be seen. Eventually, a self-replicator is able to emerge from the fray - the simplest structure capable of reliably copying its own design. If there is a random component to the self-replication process, then evolution and variation may cause this simple replicator to get more complex over time.

When it comes to universes, structures which apparently can be described by just a few parameters, such as starting mass, dimensions, and fundamental forces, simple is more common. We just can’t see all the simple universes because they are too simple to support life. And of course, by definition, we can receive no information from universes outside our own, even though we know they exist. When it comes to structures built by random processes, simplicity rules.

Move up the complexity ladder for universes, and surely they get rarer. (Unless there is a specific force manufacturing complex universes, which looks extremely unlikely. The underlying mechanism of universe creation is quantum and therefore random.) These complex universes may be rare objectively, but they are not rare to observers, who make them their home. Observers only find themselves within universes past a certain complexity level - that threshold necessary to support the emergence and flourishing of life.

There definitely exist universes more complex than our own, and we have to ask - do they contain life? In a universe with more than four dimensions, the interactions between objects would be so complex that a self-replicator would need to be fundamentally more sophisticated to launch itself, and therefore take much longer to come about by random collisions. These complex universes also must be rarer to begin with, so it seems less likely for any given intelligent observer to find him or herself within one.

This is our anthropic landscape. As typical conscious observers, we should expect to live in a universe that is simple, but not so simple that it can’t contain life. This squares nicely with our observations.

The universe’s fundamental simplicity doesn’t signify that complex forms can’t emerge within it. The human brain is coded by a portion of the genome that has less information content than Microsoft Word, but the self-similar, fractal nature of neural architecture means that an adult brain is many orders of magnitude more complex than the genome that built it.

In the same way, even if the universe in fact contains almost no information, it doesn’t mean that we can’t build things or have experiences within it that look complex to us.

For a little more on anthropics, see my October post on the self-sampling assumption, or George Dvorsky’s post from last month on the topic.

4 Responses »

The Boötes void, named after the constellation where it can be found, is the largest known region of empty space in the observable universe. Because it is so large, it is sometimes referred to as a supervoid. The void is roughly spherical and has a diameter of approximately 75 megaparsecs, or 250 million light-years, which is about 2% the diameter of the entire observable universe(!)

For comparison, our Milky Way Galaxy has a diameter of about 100,000 light-years, and the largest known galaxy is about 250,000 light years across, a thousandth the width of the void. Within this vast emptiness, only about 53 luminous galaxies have been detected, which extend in a rough tube-shape through the middle of the void. Other galaxies surely exist within the void, including structures of dark matter, but these galaxies are smaller and less massive than the 53 primaries. These 53 galaxies have an average brightness about 25% more intense than the universal average, a phenomenon that needs explaining.

Greg Aldering, an astronomer who now works at Lawrence Berkeley Laboratory, once said, “If the Milky Way had been in the center of the Boötes void, we wouldn’t have known there were other galaxies until the 1960s.” Imagine that kind of a discovery!

The void was discovered in 1981 by Robert Kirshner, Augustus Oemler Jr, Paul Schechter and Stephen Shectman in a survey of galactic redshifts. Their results were published in the paper, “A million cubic megaparsec void in Bootes” in Astrophysics Journal 248. Further studies throughout the early 80s confirmed the existence of the void, which was one of the first large voids to be detected, and is hence the most famous.

Telescopes pointing in the direction of Boötes show a sky with plenty of galaxies. What the studies showed was that all of these galaxies are either close to us or far away, with a gigantic gap in between. This gap is the Boötes void, whose center lies about 700 million light-years distant.

The Boötes void is probably the most perfect vacuum in the universe. Its density is somewhat less than that of the universe’s average, which is about one atom per cubic meter. The void’s density is certainly lower than that of typical intergalactic space, which is already extremely sparse. Like the rest of space, the most plentiful form of conventional matter to be found within the void is ionized hydrogen.

Think about it: the density of lead is 11.34 g/cm³. The density of the Boötes void is about 1.674× 10 ^ − 29 g / cm³, approximately a million million million million million times more diffuse than that. You might have heard that a neutrino of the type emitted by the Sun can be blocked by a barrier of lead about two light years across. To block such a neutrino with a barrier equal in density to the Boötes void would require the barrier to be quadrillions of times wider than the observable universe.

This extremely low density means that when a pattern of neutrinos enters in one side of the void, it looks exactly the same upon exit. Same goes for photons. Particles of matter, having much more mass than both photons and neutrinos, would of course get pulled towards the walls of the void. Because of this state-preserving property, the Boötes void may one day be seen as the ultimate time capsule - fire off a pattern of photons, only for the pattern to be rediscovered hundreds of millions of years later when it reaches the other side.

This void is so big and empty, its discovery worried cosmologists. The prevailing theory of galaxy formation suggested a more uniform distribution of matter throughout the universe, in the spirit of the cosmic microwave background radiation, which is the roughly uniform “echo” of the Big Bang. The Boötes void fell outside the range of void sizes predicted by this theory of galaxy formation. Cosmologists are still trying to create theories that better explain it.

The size of the Boötes void is too large for it to have formed as a result of natural galactic rearrangement from gravitational factors. It had to have been “set up” to be a void from the beginning of the universe. Perhaps the supervoid began as a particularly large quantum ripple on the surface of whatever exploded to become our cosmos. The Boötes void may be in that class of phenomena that requires a quantum theory of gravity to explain. Though today it is extremely large, its origins are grounded in an object many times smaller than the atomic nucleus.

There are two main theories of galaxy formation - “top-down” theories, where large structures form first and then fragment into smaller ones, or “bottom-up” theories, where large structures are formed by the coalescing of little bits. The existence of features on the scale of the Boötes supervoid is evidence in favor of the former. If the galactic formation mechanism were primarily bottom-up, the universe would be more like pudding - smooth and evenly distributed. But instead it looks more like a froth of soap bubbles, with galactic filaments separating gaping voids.

The current prevailing theory of galaxy formation is the Lambda CDM (cold dark matter) model, which is a bottom-up model.

The Boötes void is the utter absence of structure on the scale of galactic superclusters. To review, extragalactic astronomy recognizes several levels of organization of matter in the universe, beginning with galaxies, with a characteristic length of 20 kpc, which can be found within a group or cluster that measures 50 kpc to 5 Mpc across, which are further organized into superclusters with diameters topping 50 Mpc. The Boötes void is 75 Mpc across. Many other voids exist, but none this large that we have yet seen.

It is known that about 98% of the volume of the universe is consumed by intergalactic voids. The universe is made up of superclusters forming thin “walls” around these huge voids, perhaps reminiscient of the way organisms consist of cells whose main density lies in walls enclosing cytoplasm. (But in contrast to the way that atoms’ primary concentration of density is located in the nucleus.)

Of course, certain things can be found within the void, mostly in the form of energy. The extragalactic background light (EBL) and cosmic microwave background radiation (CMB) fill up the void. Interestingly, there is no gap in the cosmic background radiation in the region of the void. The void almost certainly contains a lot of dark matter and energy, perhaps even at densities no different than those found within superclusters. This means that dark galaxies and dark energy stars can probably be found. And, of course, the void is filled with endless quantities of virtual particles, which are created and annihilated constantly on the smallest timescales.

According to the Wikipedia article on the topic, the Boötes Void was mentioned in a novel by Martin Amis, “Night Train”, which centers around the mysterious suicide of a beautiful and successful astrophysicist Jennifer Rockwell. The immense size of the void leads her to conclude that there is no meaning to life, so she kills herself.

What does contemplating the size of the void lead us to conclude?

First is the logarithmic scale of human awe, which seems to be built into our brains, and is arguably a common feature in any brain built by evolution and natural selection. The Boötes Void is not that many times more impressive, subjectively, than a cave so large you can fly a helicopter around inside. Both are big, awe-inspiring, and amazing. Despite this, most people would be more impressed by the big cave, because it’s something that we can more easily imagine ourselves interacting with, something that has features similar to those we run across on a daily basis - rocks, ceilings, walls, passageways, etc. But even if you added together every cave on every planet in the universe, you’d still fall many orders of magnitude shy of the volume of that void of all voids.

It might be possible to one day build a mind that experiences a linear increase in awe with every linear increase in size, for any given void or chasm. To such a mind, a 20 cubic meter hole would be twice as impressive as a 10 cubic meter hole. I can only assume that contemplating the Boötes Void would cause this mind to self-destruct.

Secondly, the void brings to mind that everything is relative. This point is similar to the first. For example, consider a reversible agent sent into the void at relativistic speeds. A reversible agent recovers all its lost energy perfectly, so it can observe and think forever without consuming a joule. As the light of nearby galaxies slowly faded into almost complete darkness, this agent would get pretty damn bored. Most likely it would try communicating with people in the galaxies on the edge on the void, but the response time would deteriorate quickly. If the agent is lucky, it will have a very long-lived and extremely patient pen pal with which to exchange messages and play games. The agent might choose to deliberately slow down its rate of thought so that time passes more quickly. If your mind is operating slowly enough, a trip across the void might only seem like a mere decade or even a week.

Equivalently, an observer accelerated close to the speed of light experiences time more slowly. If you were accelerated to a velocity arbitrarily close to that of light speed, the trip could seem arbitrarily short. According to the Wikipedia article on the topic, a continuous acceleration of 1 g would be sufficient to allow a person to cross the entire observable universe in a subjective duration less than a normal lifetime. A similar level of acceleration would allow a traveller to cross the void quickly, but some form of braking would be advisable. At such high speeds, a typical galaxy would look like a wall of high-energy cosmic rays.

Perhaps the Boötes void will serve as a proving ground for the high-speed dragsters of the future, moving at 99.99999% the speed of light. Its extremely low density would certainly be appealing to anyone who wants to set speed records without that pesky intergalactic dust getting in the way.

A third line of speculation about the void would be the presence or absence of life within its confines. Of course, like the rest of the universe, the void is almost certainly empty, for anthropic reasons. (The probability of any given species having a neighbor is equal to the number of universes with at least two intelligent species divided by the number of universes with at least one intelligent species, which is a huge ratio.) However, if life evolved here, what would it be like? It would probably be made out of building blocks of lone hydrogen atoms floating in the void, exchanging data with photons and other quanta (gravitons?) Over an extremely long length of time, patterns of these atoms and quanta could become self-propagating, leading to a cycle of variation and evolution. However, if this were possible, it would probably happen in plain old interstellar or intergalactic space before it happened in this blank darkness. So much for that idea.

For more information, see a survey of the Boötes void. The symbolic rival of the void is arguably the Great Attractor, a cosmic aggregate of tremendous mass. For a gosh-wow narrative of the density of intergalactic space, see this article.

23 Responses »

Seeing lots of blog posts today with pictures of children looking for eggs, geeky analyses of the results of microwaving various Easter candy, et cetera, ad nauseam. Anything to avoid discussing what Easter is really supposed to be about - Jesus’ resurrection from the dead.

There were numerous active cults and sects in the Holy Land and Europe around Jesus’ time. There is no particular reason why Christianity was destined to be more or less successful than any other religious group. But it was. The values and stories of Christianity reflect what people want from a spirituality. A deeply resonant psychological archetype, possibly even developmentally predestined.

Rags to riches. A prodigy who held the attention of wise men with his philosophical and intellectual skill as a young boy. A prince whose father was not merely a king but the King of Kings. Compassionate acts of healing, even revival of Lazarus from the dead. Feeding thousands from a few loaves of bread and catches of fish. Strength in the face of temptation. A strong adherence to fundamental values, including love for one’s enemies. Betrayal and self-sacrifice. And ultimately revival from the dead and ascension into a holy realm of eternal light.

The story of Jesus of Nazareth is ultimately more about humanity and human psychology than anything having to do with incorporeal entities that even most Christians don’t take seriously anymore. Ask your Christian friends - would kindness, and brotherhood, and all these other values have meaning even if it were known for a fact that they didn’t come from a divine source? Or perhaps that good feeling we get when practicing these teachings is where the divinity comes from. We think it comes from an external source, God, when it’s actually emanating from within our brains. But some would argue that the nonexistence of God doesn’t make these values any less divine.

Jesus’ teachings, along with everyday common sense and the teachings of millions of other wise people, are part of what has been called the “human moral frame of reference”. The human moral frame of reference can’t be written down as a list of well-defined rules, but rather makes up a cluster of tendencies and feelings which we know when we see them.

We’re all a part of the same species. As such, we share far more in common with each other than with any other species, living or dead. Unless plagued by a developmental defect, we all have two arms, two legs, a pupil and a retina, a spleen and kidneys, opposable thumbs and an enlarged prefrontal cortex. There are moving parts within our brains, like the moving parts that help us choose between “right” and “wrong”, that are species-universal in exactly the same way that our other organs are.

Yes, you and George W. Bush have the same basic mental hardware for choosing between right and wrong. Part of the problem is that the human moral frame of reference is tied to the individual. But if you subtract observer-centricity, different human goal systems start looking a heck of a lot more similar. If you calibrate for access to objective facts and differences in intelligence, even more similarity emerges. In the end, the greatest objective gap in the human moral frame of reference is probably between men and women, who have brains that differ slightly on a neurological level, not between any two men or any two women. But still, in comparison to every other intelligent species that can theoretically exist, all humans - men and women alike - basically have the same underlying hardware used to make moral choices.

This species-universality of morality has made some people hopeful that we will be able to successfully build an intelligent machine that shares humanity’s basic values, without being biased in favor of the programmers. It turns out that this could probably be possible even if several different intelligent species lived together on earth, but all being members of the same species sure doesn’t hurt.

For people thinking through the necessary features of a Friendly AI, the question to ask is not “what separates humans from each other?” but rather, “what do we fundamentally have in common?” This question is the province of cognitive psychology and social science, not politics or water-cooler gossip. We need to have a rich, information-theoretic description of human goal systems so we can teach these newcomers which changes we’d see as good and which we’d see as bad. They might seem obvious to us, but for an entity starting with literally zero cognitive content, the stretch to even a basic form of morality is a long one.

Does this line of inquiry interest you? Try continuing with the eminently readable Dialogue on Friendliness, which explores fundamental issues surrounding morality and goal systems.

3 Responses »

In the news this week is something extremely relevant to last week’s discussion on nanofactory regulation - on Monday IBM announced a computer security technology that encrypts data at the hardware level, something which has been done before but apparently not to this extent. The technology is called SecureBlue. A quote from news coverage on my.freeze.com:

There are multiple ways to achieve encryption, the mathematical art of encoding data to protect it from spying eyes. Specialized software can do the trick, as can hard-wired chips inside computers.

But IBM researchers contend that unless the encryption function is performed by a computer’s central processing unit, a supremely savvy hacker can tap into the pathway between the machine’s brain and the separate encryption engine.

To guard against that, IBM is announcing Monday that it has developed “SecureBlue” a set of encryption circuitry that can be integrated into any processor, regardless of its manufacturer.

“This thing is trying to be one of the most paranoid devices on the planet,” said Charles Palmer, IBM’s head security researcher.

One of the only times data is not encrypted within a SecureBlue chip is when it is displayed on the screen. We are rapidly approaching an era where the weakest link in computer security will almost always be the user (if we aren’t there already). IBM sees the technology being used in a variety of devices from PCs to handhelds and beyond.

The encryption scheme is not computation-heavy, barely consuming any overhead. IBM seems confident that the basic security of the technology will hold even if a hacker has ways of intimately monitoring the data streams within the hardware. Presumably access to IBM’s proprietary security algorithms is the only way to crack the code - and I would expect that the details of these algorithms are only known to at most a few hundred (more likely a few dozen) people at IBM’s research labs. Even with these algorithms, it’s not certain it would even be possible to decode any given package of encrypted data because it would be associated with a long, randomized string of bits (key). Perhaps the technology has safeguards so that data can be recovered even if the keys are lost? If so, a list of keys might be kept in a centralized location managed by IBM.

SecureBlue might be seen as a complement or successor to Trusted Computing.

It is fortuitous that IBM is the source of this new security technology, as it is also the company that built the computer that almost beat the highest-rated chess player on earth (Deep Blue), the fastest computer on earth (Blue Gene), and the largest attempt at a computer simulation of the mammalian brain (Blue Brain).

Hollywood and the big business behind proprietary software will be cheering for this technology, because it gives them another way to potentially prevent consumers from copying their movies, music, software, etc. I’m cheering it on for slightly different reasons, that is, the technology’s role in protecting us from future risks associated with totally unrestricted computers and software.

Sometime in this decade or the next, there will be a revolution in desktop manufacturing. This needn’t be in the form of nanofactories - it may debut as a relatively expensive machine that uses macroscale technology to shape plastic and electronics components into toys, tools, and simple gadgets. People will eventually be able to make custom products of high quality in their own home for low cost. The revolution is already starting to happen, with machines like MIT’s “fab lab” and the MCP Realizer, among dozens of others.

When copyrighted media such as songs or video clips get duplicated and distributed, the recording and film industries take a big hit. When expensive software like AutoCAD, Quickbooks, Windows XP, Photoshop, and Maya are copied, the software industry takes a big hit.

But these might be ultimately unavoidable. Information has a tendency to run free, and if security isn’t built into the foundations of the technology, it’s futile to stop the torrent by suing people one by one.

With desktop manufacturing, it will be a different story. If the design for a hot new product becomes public knowledge, then the value of the product will plummet mere days after its release, eliminating the motivation to both invent and invest. The dangers of a malfunctioning product will potentially be duplicated millions of times over. Without hardware-level restrictions and safeguards, performing a recall on a home-fabbed product will be neigh impossible. To make things worse, copycats will attempt to create similar products that circumvent safety restrictions.

This is just the beginning of why hardware-level encryption will be so much more important in the future than today. There are large classes of both existential risk and intense global nuisance that will be facilitated by insecure computing. These include bioweapon design, missile design, cyber-terrorism, remote control of military hardware, and much more.

The single greatest long-term risk of powerful insecure computing is probably self-improving Artificial General Intelligence (AGI) that is indifferent to human welfare. As available computing power increases, it gets easier to build an intelligent computer. (How much easier we don’t know.) However, it doesn’t get any easier to build an intelligent computer that cares about humans with the same complexity and subtlety that we care about each other - a must if you’re aiming for smarter-than-human AI. To oversimplify a bit, the former is a matter of trying things out until something works, and the latter is about developing a formal theory of what an agent will do given a starting set of preferences and the ability to reprogram itself recursively. Both will take a lot of brains, but creating any AI is a problem that lends itself to brute forcing much more than creating a certain type of AI.

Before we let powerful, unrestricted computers be available to just anyone, we should solve the problem of Friendly AI. A successful solution would give us allies who actually grew up in the world of code and will have a much better idea of which types of computation are truly dangerous and which are harmless - a question which humans are ultimately unqualified to answer.

8 Responses »

The Singularity concept got a nice dose of publicity yesterday, with Instapundit blogebrity and soon-to-be New York Times columnist Glenn Reynolds coming out strongly in favor of the Singularity. (For those of you not in the know, Reynolds is kind of a big deal - his blog scores a million visits per week.) Like other big media people when they get their teeth on the idea, it begins with the standard Vingean definition (technological creation of smarter-than-human intelligence) then phases into the broader, technology-centric Kurzweilian definition (guarantees of extended life, peaceful fusion with machines, etc.) This is where Singularitarians like to come in and repeat the mantra: the Singularity is about greater intelligence, not more technology. Greater intelligence will likely end up inventing and applying new technologies for a variety of ends, but thinking of the Singularity primarily as “really cool tech” misses the point.

The Singularity is about breaking through the millennia-enduring glass ceiling of Homo sapiens cognitive capacity. There are people that can bench press 500 lbs, and people that can barely manage 100. There are atheletes that can run a five-minute mile, and there are geeks that can puff through a mile in under ten minutes. There are people that do AIDS research and there are people that couldn’t tell you what a cell is.

But there is nobody who can bench press twenty tons. There is no one that can run a mile in two minutes. And you’d be hard pressed to find someone that can take one look at the details of the AIDS problem, and have a full cure within a couple weeks. All these tasks represent abilities, and there are winners and losers for each ability. But the fact that we all belong to the same species puts us in the same general category of competency, no matter what ability you choose. Sometimes it is hard to realize this, because we only have other humans to compare ourselves with.

A naive human might look at a champion weightlifter and think that he’s seeing the best there is and the best there ever will be. But think - if we evolved on a planet with higher gravity, all of us might have stronger muscles, and then bench pressing 500 lbs would simply be average. Scientists have developed artificial muscles 100 times stronger than those we possess.

If we created a cyborg with artificial muscles and had this strongman lift weights, could we predict the result?

Largely, yes. He’d be able to lift a hell of a lot of weight. But the little quirks that would come from superhuman strength are the sort of thing that excites scriptwriters and fans of science fiction and comics - maybe the cyborg would get from place to place by jumping instead of walking, for example. Or maybe he would move his furniture solo rather than with a four-man crew. The fact of the matter is, we wouldn’t be able to predict exactly how a superstrong human would live or act - we’d have to actually create one and see how it got through life with his or her own unique perspective.

The uncertainty would be far more explosive if we had a means of enhancing human intelligence. It would be far more difficult to predict how a superintelligence would think or act than to predict the actions of a superman. A superman would still have human needs, wants, desires, emotions, thinking habits, flaws, and motivations. A superintelligence, especially in the form of an AI, could have completely rearranged desires and ways of achieving them. It could circumvent restrictions placed on it through sheer cunning, cleverness, and social engineering. A superintelligence might be able to walk into a lab and point out ten different things the resident scientists never thought of, accelerating research by years or even decades. You wouldn’t be able to predict the moves a superintelligence would choose in a chess game, though you could predict it would win.

That is what the Singularity is about. The only reason that technology so frequently enter into the discussion is that technology is the only way we know of to plausibly create a superintelligence. Characteristic of Singularity discussions is mentioning the feedback effect - smarter minds would be better at upgrading their own intelligence. Imagine if, rather than technology, we used magic to manipulate the external world. Then we would create superintelligence with a particularly powerful magic spell. The superintelligence would then invent more incantations and sorceries, and use them to further expand her magical repertoire. Perhaps in a parallel universe, this is how Singularities happen.

The Singularity discussed by folks like Glenn Reynolds and Ray Kurzweil is more about the next stage of technology than it is about the next stage of insight, intelligence, wisdom, cleverness, sociality, or morality. Singularitarians are merely looking at technology as the only available means to a very worthy end - intelligence smart enough to improve its own intelligence recursively. So next time you see the words “technology” and “Singularity” in the same article, remember that there are people out there calling themselves singularitarians who don’t really care about shiny gadgets all that much. We just want to explore the potential for smarter, more benevolent types of thinkers.

What reading this Glenn Reynolds article does do is show us that transhumanism - not singularitarianism - is becoming radically more mainstream. And a particularly blasphemous form of transhumanism as well. Reynolds talks about…

Limitless lifespans, if not immortality, superhuman powers, virtually limitless wealth, fleshly pleasures on demand, etc.

This is pretty serious stuff to be discussing on a level-headed website like Tech Central Station. And it certainly would look pretty insane in a New York Times column. More:

Reynolds’ article defends transhumanists and other secularists who see mankind’s future lying in technological and scientific advancement rather than Heaven and Jesus. The classic knee-jerk reaction to transhumanist ideas for the unexposed - “the whole thing is a damn cult”. Not only does he dismiss assertions that the Singularity or transhumanism is a religion, but states that, in fact, the prospect of achieving things we’ve wanted throughout history through technological means is something to embrace and be excited about, rather than write off.

Christianity talks about achieving eternal life, joining a perfect society, and experiencing endless happiness by being good people on earth. Standard atheism is the cynic’s answer - “you guys are all dreaming, but we’re savvy intellectuals who know that life actually sucks and after death there is only rotting of the body”. It’s cool to be a cynic. It’s easier to demolish a complex argument than create a well-supported one from scratch.

Then along comes the prospect of serious life extension. Nanotechnology that makes products for the cost of their raw materials. Cyborgs flying around and seeing with ultraviolet vision. Etc.

The atheists cry bullshit, because it sounds too similar to the Christian doctrine that originally let them down intellectually. But these are real prospects - and they have no right to dismiss them off-hand without serious investigation.

But high-profile people like Reynolds, who end up writing articles like this, are telling these people to give the ideas another look. Reynolds seems to be coming out of the closet as a transhumanist in a feisty way, and this is just as he’s going to work for the New York Times.

Some people, like venemously anti-transhumanist journalist John Bruce, are terribly unhappy about this. The guy writes a blog with post after post attacking Reynolds, and whining about his pet peeve - cryonics. (On the lighter side of things, Bruce’s hobby is working with model trains.) Conservative legislators in Missouri are even kicking around ideas for an amendment to the state constitution, “Regulation of Human-Animal Crossbreeds, Cloning, Transhumanism, and Human Engineering Is Reserved to the People”. These two pages are highly LOL-worthy material, and I encourage you to look into them. I’m sure that John Bruce would especially appreciate your thoughtful feedback on his posts.

On the more futurism-friendly side of the blogosphere, transhumanist Phil Bowermaster covered the article, and added his thoughtful analyses on the issue. He has also opened a Singularity survey, which you should participate in, for great justice. Post your answers here if you’d like.

Thanks to big boys like Reynolds coming out for us little fringe futurists, we will surely be free to talk about omnipotent transhuman AIs without accusations of religiosity. Everyone knew we were entirely secular to begin with, right? Not a smidgin of cultishness to be found, as any well-educated gentleman or gentlewoman can testify.

12 Responses »

As part of my participation on the CRN Task Force, I’ve been thinking in a bit more detail about how nanofactories might be regulated in the future. I recently posted the following in some forums. Join the disussion if you like. This was prompted by the questions, “What guidelines for humanity’s use of nanotechnology should there be? How should they be enforced? I see nanotech changing the sociology of the world as much or more than it will change the economy of the world.”

~~~~~~~

Nanotechnology should only exist for public use in the form of personal nanofactories (PNs), self-contained desktop units that manufacture products associated with a license.

PNs will only fabricate product designs signed by a safety authority. Products will have a maximum and minimum allowable size, chemical composition, and power consumption, based on their documented functions. Other limitations to product designs should be applied based on the recommendations of an expert committee.

Because of physical scaling laws that permit extreme productivity at the nano-scale, the first company to create nanofactories will be able to manufacture their product rapidly. Once the productivity and flexibility gains inherent to desktop nanomanufacturing become obvious, there will be instantaneous and sustained worldwide demand.

The business model

Nanofactories should be sold for a reasonable price, perhaps similar to the debut price of state of the art computers (~$3,000), with a pricing half-life of three months. The technology itself theoretically allows a pricing half-life of a day or so, but a lengthy rollout will be used to generate enthusiasm and work out the bugs. The system will be $200 in a year and $10 in two years. Or maybe there will be a natural price floor. Note that this pricing model artificially slows the rate of adoption by a factor of about 100. This allows feedback to flow on normal human timescales, allowing discussion and analysis of potential problems before they happen. The pricing model may not happen in reality, but it’s a solid possibility.

The company should take their product international at a balanced pace - slow enough to work out the bugs, fast enough to ensure that they stay on top of the competition. Nanofactories are a sufficiently revolutionary technology that the first mover should be able to gain global dominance through competitive pricing and intelligent acquisitions. A unified company in charge of nanofactories will also simplify policy issues and guarantee the enforcement of universal security and safety standards.

The real price of the nanofactory will be in the products. Third-party developers will use an API provided by the nanofactory company to design and license products. Licensed products are sold to customers through an interface on the nanofactory. Pricing tiers for multiple product copies will be set by product developers. The nanofactory company will grab a small percentage of the profit, but most will go to the developers.

The API will be a CAD system that allows designers to specify high-level characteristics of an object or system without knowing its details on the molecular level. Drop-n-drag interfaces will allow anyone to design simple products. Developers need to be given flexibility such that they can let their minds run free, without feeling the limitations of a proprietary platform. This will be achieved by the design of nano-blocks by the nanofactory company - verified modular components that simulate surfaces or materials, store and transmit electricity, light, or force, communications cables and processors, displays and interfaces, and much much more.

The data underlying the operation of nanofactories will not be open-source or reprogrammable. It will, however, be reviewed and continually redesigned by the brightest engineers and security experts in the industry. Made impervious to natural disasters, internal scanning, and reverse-engineering, nanofactories will always keep records of who is using them, their respective energy budgets, local and global laws, and library of manufacturable products. These desktop machines will have hundreds of terabytes of hard drive space for storing fabrication instructions and product designs. Their tamper-proof nature will allow nanofactories to serve as ideal “black boxes” to examine after disasters, both natural and artificial. Conversely, nanofactories should be programmed to fry their internal workings when they recognize they are being breached. An opaque “airlock” should prevent the product output port from serving as a window to scanning the nanofactory’s internals from the outside.

Law

A primary concern for the development of civilian and commercial nanofactories is the buildup of NanoTrash - cheaply mass-manufactured products made of mostly diamond and empty space. Avoiding NanoTrash while preserving our freedom to design and create will be a great challenge of the early nanotech era. For starters, each nanofactory user should have a personal matter and energy budget determined by a safety authority. These limits should be variable based on product class and user profession. For example, someone that works at a hospital should have a larger energy budget when it comes to manufacturing medical products. In the same way that it’s illegal for just anyone to randomly practice medicine, not just anyone should be permitted to manufacture large quantities of painkillers, syringes, or scalpels.

Many professions operate under licenses today. Physiotherapists, acupuncturists, emergency medical technicians, paramedics, doctors, nurses, teachers, lawyers, and professional engineers all require some form of licensure to work their jobs. These licenses represent that the licensee demonstrates basic knowledge of their profession and its associated responsibilities. Because of the tremendous range of products nanofactories will make available cheaply, licensing and energy budgets are a must to ensure that dangerous products do not fall into ignorant or malicious hands. Crowd-mediated reputation markets will quickly label the black and white hats in the fabrication business, leaving massive paper trails for both law enforcement and avid groupies. Try to model and fabricate a torture device, and certain design privileges are temporarily suspended. Design a useful product, and you are rewarded with an increased energy budget.

The most widely-used and largest products will have the highest energy budgets - storage containers, automobiles, housing, civil systems, renewable power plants, and agricultural tools. Nano-built products will quickly outperform and underprice variants manufactured using older technologies. The rate at which this occurs will depend upon the improvement of the underlying nanofactory/API technology. The most necessary, universal, and algorithmically simplest products will be the first to be ported to nanofactories. Products containing any subset of the nanofactory technology itself (actuators, computers, sensors, purifiers, other electronics and structural elements) will become immediate candidates for design and licensing.

Because they will be competing for the finite energy budget of the consumer, firms designing new products will have a reason to care. Successful firms will be granted larger energy budgets and perhaps even greater design flexibility.

Complex organic products like food will not be built by the early, all-diamond nanofactories. In fact, anything that can’t be made exclusively from diamond will continue to be produced by traditional industries. But if I look around my house, it’s difficult to find objects that can’t theoretically be made from diamond - okay, maybe blankets, stuffed animals, mirrors, pasta, pineapples, and certain clothing can’t be made out of diamond, but what can? But how about stoves, tables, chairs, televisions, lamps, lampshades, clocks, computers, storage units, pots, pans, walls, locks, and boxes? Quite likely. Most diamondoid products will be made of 99% air or vacuum and will be ballasted by water vapor, but they will serve their function.

Quick and efficient recycling for nanotech products is a must. A large household storage tank for temporary or permanent deposit or withdrawal of water, carbon, and other feedstock or byproducts will quickly become universal. This will be built into a new civic infrastructure.

All products will be fabricated with multiple inbuilt copies of its signed safety certificate and an associated key - a simple “watermark” that lets law enforcement know the legal status of the product while ensuring that product designers get to collect their well-deserved licensing fees. If a product is found to be dangerous, the associated key is revoked, and the product is either deactivated remotely or added to a warning list. To enable the immediate deactivation of any dangerous product, designs should incorporate emergency shutdown features that respond to broadcasts of revoked safety keys.

Nanotech economy

After the initial wave of diamondoid products will come new functionality - traditional materials like simulated wood, metal, ceramic, stone, and the huge polymer family, which includes all plastics. These optimized materials might be made of different molecules than the originals but will offer superior performance and safety while consuming fewer resources.

Investors will balk at the falling prices of real estate, raw materials, energy, and just about everything. To ensure that runaway hyperdeflation does not occur will require a minimum price tag per kilogram of product in conjunction with a personal energy budget. Then comes the question - should everyone on Earth have the same allotted energy budget, or should it vary based on salary, education, productivity, reputation, honesty, or some other characteristic?

The world may need to make a choice between pure democracy and simple survival. To preserve the status quo and maximize continuity with the past, some system based on a combination of money in the bank and credit may be chosen. Or perhaps something more modern, like PageRank, where engineers and designers are assigned budgets based on referrals. Or a system closely tied to attention like Alexa.com, where the engineer’s designs are judged based on the number of people aware of them - their “reach”. Or, perhaps most appropriately, an unbiased, flexible inference engine that assigns projects preference and resources based on sophisticated volition-extrapolation models of every human individual. Science-fictional-sounding maybe, but nanotech will make it possible.

In order for the human race to go on, it needs to survive the nanotech era without too large of a disaster. As such, the probability of disaster should always be kept below a certain threshold. Varying acceptable disaster probability thresholds (DPTs) will ostensibly be voted upon by communities at the local, regional, national, and international levels. Ideally, different regulation sets will correspond to known DPTs. Stringent regulations minimize the probability of disaster, lenient regulations increase it.

Unfortunately the analyses underpinning these thresholds will no doubt be politically sensitive and value-laden. Because nanofactory issues will be global in scope, there will be strong pressure towards the interaction and unification of political parties across national lines. A global political party or even government could emerge.

Potential dangers will come from several main categories - chemical, biological, nuclear, and physical. Nanotech will deepen all these threats and magnify nascent dangers such as electromagnetic and virtual weaponry. Electromagnetic dangers will include satellite-based microwave beams and other forms of lethal and non-lethal directed energy. Virtual weaponry will include advanced Artificial Intelligence, robotics, and decision support systems.

We may see that the more dangerous products can be defined in terms of complexity rather than by size or energy consumption. This may lead to a personal complexity budget alongside an energy/matter budget.

The speed of new computers will be considered a problem. Extremely fast supercomputers running arbitrary code is dangerous. As a result, supercomputers should only run code signed by a safety authority. Computer scientists should require licenses to operate or manufacture supercomputers above a certain speed limit. This speed limit should be only slightly past the limits of conventional semiconductor technology - this is something like 100 times the power of today’s computers, and should be sufficient for most purposes. Computers built using new operating principles, such as plasmonics, photonics, and DNA computing should also respect this speed limit. Quantum computing may prove hard to enforce limitations with. If it turns out that quantum computing allows rogues to easily design and simulate virtual or physical weapons, it could end up being forbidden altogether.

Personal computers should also have speed limits, also hovering slightly past the limit of conventional semiconductor technology. Hobbyists who desire crunch power for special projects will fabricate special-purpose computers that are only mechanically able to run certain safety-verified algorithms.

Breakthrough technologies

Even though nanotechnology itself is a revolutionary technology, it could give rise to even greater ways of controlling the structure of matter. This includes cybernetics, including human intelligence augmentation, and Artificial Intelligence. Limits should be placed on products designed to modify human biological characteristics. For ethics and safety reasons, steps in this direction should be taken slowly and carefully. Worldwide enforcement of these standards are a must. Because the proprietary nanofactory technology will presumably be universal, standard enforcement will be feasible if the machine is tamper-proof and only builds approved products.

If the world continues to be democratic in nature, the prospect of accelerating the human birth and growth cycle (”nano-fertility”) could present itself as a strategy for certain countries or cultures to tip the scales in their favor. As a result, products designed to accelerate the cycle of pregnancy should be forbidden. The prospect of life extension (which should be regulated less stringently than pregnancy) will prompt the setting of a child limit in the civilized world - something like two or three children per couple. As women are educated, contraception becomes universal, and manual labor is less valued, developing countries will follow the trend.

The most interesting (and potentially destabilizing) prospect of cybernetics and Artificial Intelligence is the potential to create geniuses from average individuals using enhancement procedures, or to create human-rivaling AI from scratch. These forays should be limited until safety studies determine the best angle of approach.

20 Responses »