NS: Rise of the replicators

Enthusiasts are building machines that can make just about anything – including their own robotic offspring. New Scientist explains:

Still, ingenious as these machines are, they merely churn out piles of parts. What about assembly? A heap of plastic and metal is not a machine, just as you don't have much in common with a pile of flesh and bones.

Greg Chirikjian, a roboticist at Johns Hopkins University in Baltimore, Maryland, agrees. "When a prototype only makes parts, the machine that made those parts wasn't reproduced," he says. A true self-replicator must handle both fabrication and assembly. Chirikjian and his colleague Matt Moses are aiming to achieve this with a kind of Lego set that doesn't need anyone to play with it.

The pair have already demonstrated key parts of such a system, using around 100 plastic blocks. Although it cannot yet fabricate these blocks itself, the machine is able to move in 3D to pick up and bind them into larger structures. Moses is currently working on having it make a complete replica of its own structure using Lego-like bricks, though the machine still relies on conventional motors - which have to be installed by hand - to drive its activity.

Ranking the most powerful forces in the Universe

There are a large number of forces at work in the Universe, some more powerful than others -- and I'm not talking about the four fundamental forces of nature. A force in the context I'm talking about is any phenomenon in Universe that exhibits a powerful effect or influence on its environment. Many of these phenomenon quite obviously depend on the four basic forces to function (gravity, electromagnetism, the weak interaction and the strong interaction), but it's the collective and emergent effects of these fundamental forces that I'm interested in.

And when I say power I don't just mean the capacity to destroy or wreak havoc, though that's an important criteria. A force should also be considered powerful if it can profoundly reorganize or manipulate its environment in a coherent or constructive way.

Albert Einstein once quipped that the most powerful force in the Universe was compound interest. While he does have a point, and with all due respect to the Master, I present to you my list of the four most powerful phenomenon currently making an impact in the Universe:

4. Supermassive Black Holes

There's no question that black holes are scary; it's the only part of the Universe that can truly destroy itself.

Indeed, Einstein himself, whose Theory of Relativity opened the door to the modern study of black holes, noted that "they are where God has divided by zero." And it's been said that the gravitational singularity, where the laws of physics collapse, is the most complex mystery of science that still defies human knowledge.

Somewhat counterintuitively, black holes take the weakest of the four basic forces, gravity, to create a region of space with a gravitational field so powerful that nothing, not even light, can escape its pull. They're called "black" because they absorb all the light that hits them and reflect nothing. They have a one-way surface, the event horizon, into which objects can fall, but out of which nothing (save for Hawking Radiation) can escape.

Black holes can also vary in size and gravitational intensity. Supermassive black holes are a million to a billion times the mass of a typical black hole. Most galaxies, if not all, are believed to contain supermassive black holes at their centers (including the Milky Way).

And recent studies are now suggesting that they are much larger than previously thought. Computer models reveal that the supermassive black hole at the heart of the giant galaxy M87 weighs the same as 6.4 billion suns—two to three times heavier than previous estimates.

That's a lot of pull.

Indeed, should anything have the misfortune of getting close enough to a supermassive black hole, whether it be gas, stars or entire solar systems, it would be sucked into oblivion. Its gravitational pull would be so overwhelming that it would hurl gas and stars around it at almost the speed of light; the violent clashing would heat the gas up to over a million degrees.

Some have suggested that the supermassive black hole is the most powerful force in the Universe. While its ability to destroy the very fabric of space and time itself is undeniably impressive (to say the least), its localized and limited nature prevent it from being ranked any higher than fourth on my list. A black hole would never subsume an entire Galaxy, for example, at least not within cosmologically long time frames.

3. Gamma-Ray Bursts

The power of gamma-ray bursts (GRB) defies human comprehension.

Imagine a hypergiant star at the end of its life, a massive object that's 150 times larger than our own. Extremely high levels of gamma radiation from its core is causing its energy to transform to matter. The resultant drop in energy causes the star to collapse. This results in a dramatic increase in the thermonuclear reactions that was burning within it. All this added energy overpowers the gravitational attraction and it explodes in a fury of energy -- the hypergiant has gone hypernova.

This is not the stuff of fiction or theory -- explosions like this have been observed. Hypernovas of this size can instantly expel about 10X46 joules. This is more energy than our sun produces over a period of 10 billion years. 10 billion years! In one cataclysmic explosion!

Hypernovas can wreak tremendous havoc in its local area, effectively sterilizing the region. These explosions produce highly collimated beams of hard gamma-rays that extend outward from the exploding star. Any unfortunate life-bearing planet that should come into contact with those beams would suffer a mass extinction (if not total extinction depending on its proximity to the supernova). Gamma-rays would eat up the ozone layer and indirectly cause the onset of an ice age due to the prevalence of NO2 molecules.

Supernovas can shoot out directed beams of gamma-rays to a distance of 100 light years, while hypernovas disburse gamma ray bursts as far as 500 to 1,000 light years away.

We are currently able to detect an average of about one gamma-ray burst per day. Because gamma-ray bursts are visible to distances encompassing most of the observable Universe -- a volume encompassing many billions of galaxies -- this suggests that gamma-ray bursts are exceedingly rare events per galaxy. Determining an exact rate is difficult, but for a galaxy of approximately the same size as the Milky Way, the expected rate (for hypernova-type events) is about one burst every 100,000 to 1,000,000 years.

Thankfully, hypergiant Eta Carinae, which is on the verge of going nova, is well over 7,500 light years away from Earth. We'll be safe when it goes off, but you'll be able to read by its light at night-time.

But not so fast -- our safety may not be guaranteed. Some scientists believe that gamma-ray busters may be responsible for sterilizing giagantic swaths of the galaxy -- in some cases as much as a quarter of the galaxy. Such speculation has given rise to the theory that gamma-ray bursters are the reason for the Fermi Paradox; exploding stars are continually stunting the potential for life to advance, making it the 3rd most powerful force in the Universe.

2. Self-Replication

A funny thing started to happen about 8 billion years ago: pieces of the Universe started to make copies of itself. This in turn kindled another phenomena: natural selection.

While this might not seem so impressive or powerful in its own right, it's the complexification and the emergent effects of this process that's interesting; what began as fairly straight forward cellular replication, at least on Earth, eventually progressed into viruses, dinosaurs, and human beings.

Self-replicating RNA/DNA has completely reshaped the planet, its surface and atmosphere molded by the processes of life. And it's a process that has proven to be remarkably resilient. The Earth has been witness to some extremely calamitous events over its history, namely the Big Five Mass Extinctions, but life has picked itself up, dusted off, and started anew.

Now, what makes self-replication all the more powerful is that it is not limited to biological substrate. Computer viruses and memes provide other examples of how self-replication can work. Replicators can also be categorized according to the kind material support they require in order to go about self-assembly. In addition to natural replicators, which have all or most of their design from nonhuman sources (i.e. natural selection), there's also the potential for:

• Autotrophic replicators: Devices that could reproduce themselves in the wild and mine their own materials. It's thought that non-biological autotrophic replicators could be designed by humans and could easily accept specifications for human products.
• Self-reproductive systems: Systems that could produce copies of itself from industrial feedstocks such as metal bar and wire.
• Self-assembling systems: Systems that could assemble copies of themselves from finished and delivered parts. Simple examples of such systems have been demonstrated at the macro scale.

It's conjectured that a particularly potent form of self-replication will eventually come in the form of molecular manufacturing and the introduction of self-replicating nanobots. One version of this vision is connected with the idea of swarms of coordinated nanoscale robots working in tandem.

Microscopic self-replicating nanobots may not sound particularly powerful or scary, but what is scary is the prospect for unchecked exponential growth. A fear exists that nanomechanical robots could self-replicate using naturally occurring materials and consume the entire planet in their hunger for raw materials. Alternately they could simply crowd out natural life, outcompeting it for energy. This is what has been referred to as the grey goo or ecophagy scenario. Some estimates show, for example, that the Earth's atmosphere could be destroyed by such devices in a little under two years.

Self-replication is also powerful in terms of what it could mean for interstellar exploration and colonization. By using exponentially self-replicating Von Neumann probes, for example, the Galaxy could be colonized in as little as one to ten million years.

And of course, if you can build you can destroy; the same technology could be used to sterilize the Galaxy in the same amount of time [for more on this topic read my article, "Seven ways to control the Galaxy with self-replicating probes"].

Consequently, self-replication sits at #2 on my list; its remarkable ability to reshape matter, adapt, grow, consume, build and destroy make it a formidable force to be reckoned with.

1. Intelligence

Without a doubt the most powerful force in the universe is intelligence.

The capacity to collect, share, reorganize and act on information is unlike anything else in this universe. Intelligent beings can build tools, adapt to and radically change their environment, create complex systems and act with reasoned intention. Intelligent beings can plan, solve problems, think abstractly, comprehend ideas, use language and learn.

In addition, intelligence can reflect on itself, predict outcomes and avoid peril; autonomous systems, for the most part, are incapable of such action.

Humanity, a particularly intelligent bunch owing to a few fortuitous evolutionary traits, has -- for better or worse -- become a force of nature on Earth. Our species has reworked the surface of the planet to meet its needs, significantly impacting on virtually every other species (bringing many to extinction) and irrevocably altering the condition of the atmosphere itself. Not content to stay at home, we have even sent our artifacts into space and visited our very own moon.

While some cynics may scoff at so-called human 'intelligence', there's no denying that it has made a significant impact on the biosphere.

Moreover, what we think of as intelligence today may be a far cry from what's possible. The advent of artificial superintelligence is poised to be a game-changer. A superintelligent agent, which may or may not have conscious or subjective experiences, is an intellect that is much smarter than the best human brains in practically every field, including problem solving, brute calculation, scientific creativity, general wisdom and social skills. Such entities may function as super-expert systems that work to execute on any goal it is given so long as it falls within the laws of physics and it has access to the requisite resources.

That's power. And that's why it's called the Technological Singularity; we have no idea how such an agent will behave once we get past the horizon.

Another more radical possibility (if that's not radical enough) is that the future of the Universe itself will be influenced by intelligent life. The nature of intelligence and its presence in the Universe must always be called into question. There exists only one of two possibilities: intelligence is either 1) cosmological epiphenomenon, or 2) an intrinsic part of the Universe's inner workings. If it's the latter, perhaps we have some work to do in the future to ensure the Universe's survival or to take part in its reproductive strategy.

Theories already exist in regards to stellar engineering -- where a local sun could be tweaked in such a way to extend its lifespan. Future civilizations may eventually figure out how to re-engineer the Universe itself (such as re-working the constants) or create an escape hatch to basement universes. Thinkers who have explored these possibilities include Milan Cirkovic, John Smart, Ray Kurzweil, Alan Guth and James N. Gardner (for example, see Gardner's book Biocosm: The New Scientific Theory of Evolution: Intelligent Life is the Architect of the Universe).

Intelligence as a force may not be particularly impressive today when considered alongside supermassive black holes, gamma-ray bursts and exponential self-replication. But it may be someday. The ability of intelligence to re-engineer its environment and work towards growth, refinement and self-preservation give it the potential to become the most powerful force in the Universe.

Latest podcast posted: 2008.03.11

The latest episode of the Sentient Developments Podcast is now available. Alternative audio formats are also available.

In this episode I discuss seven ways to control the Galaxy with self-replicating probes, the problem with 99.9 % of so-called 'solutions' to the Fermi Paradox, and why we should work to overcome gender.

Seven ways to control the Galaxy with self-replicating probes

So, you want to take over the Galaxy. A good career move. Ultimately, you're hoping to communicate with extraterrestrials, colonize entire sets of star clusters, and eventually lord it over the entire Milky Way.

You've got the motive, but what about the means?

Well, forget about generation ships, suspended animation or ringworlds – the best way for you to explore, colonize and ultimately rule the Milky Way will be through the use of self-replicating robotic spacecraft – what are sometimes referred to as von Neumann probes.

Von Neumann's idea

Back in late 1940’s the brilliant mathematician John Von Neumann wondered if it might be possible to design a non-biological system that could replicate itself. Von Neumann wasn’t thinking about space exploration at the time, but other thinkers like Freeman Dyson, Eric Drexler, Ralph Merkle and Robert Freitas later took his idea and applied it to exactly that.

The strength of Von Neumann's idea lies in the brute efficiency of exponential growth. Given enough time and patience, a single self-replicating probe could produce millions upon millions of offspring; it would be like a massive bubble expanding outward into the Galaxy. It’s possible that these probes could come to occupy all four corners of the Milky Way in as little as half a million years – even if each probe travels at an average cruising speed of one tenth the speed of light.

In order to work, however, a von Neumann spacecraft would have to be put together using advanced nanotechnology and artificial intelligence -- technologies that we have yet to develop. In fact, the device itself would be a molecular assembler, capable of reconstituting matter into copies of itself.

A number of scientists and sci-fi writers have speculated over the years about the different kinds of probes we might want to construct once we're ready to explore space in this fashion. Other thinkers, namely astrosociobiologists, have wondered if extraterrestrials have constructed probes of their own.

I recently took a look at these visions and came up with a Von Neumann probe taxonomy. I came up with 7 basic spacecraft functions:

1. Exploration
2. Communication
3. Working
4. Colonization
5. Uplifting
6. Berserking
7. Policing

These tasks don’t have to be exclusive to a single probe. It’s possible that probes will be fairly versatile, able to change their functions as circumstances dictate. That said, you're likely going to need all these probes in your effort to take over and control the Milky Way.

Here’s how the different probes will work:

1. Exploration probes

These probes would be designed strictly for space exploration and surveillance; they would not contact or interact with other intelligent civilizations. We have already created such probes, namely Voyager 1 and 2 – although strictly speaking they are not von Neumann replicators.

Exploration probes could remain local and explore our Solar System (what has been dubbed Astrochicken probes), or they could be sent on interstellar missions to explore and transmit their findings back to Earth.

Admittedly, the timescales in question are significant – at least to modern human lifespans and our reasonable expectations for return on investment. But the information these probes could provide would be invaluable. They could study foreign solar systems in exquisite detail – and even alert us to the presence of extraterrestrial life.

These probes could also act as stationary reconnaissance stations. They could take residence in a data rich area and continuously beam that information back to Earth--all without ever being detected.

2. Communication probes (a.k.a. Bracewell probes)

The current SETI strategy of targeting stars and listening for radio signals has an extremely slim chance of success. It’s a needle-in-the-haystack approach. That said, given the assumption that civilizations want to communicate with us, a more efficient way for them to make contact would be to disseminate self-replicating communication probes across the Galaxy.

Dubbed Bracewell probes (named after Ronald N. Bracewell who thought of the idea back in 1960), these devices would work as an alternative to interstellar radio communication between widely separated civilizations. This strategy only makes sense given the inefficiency and weakness of radio signals emitted from the source planet.

Christopher Rose, an electrical engineer at Rutger’s University, has suggested that we should actually look for these probes in our own Solar System. He argues we should be checking the mail instead of waiting for a phone call.

Multiple Bracewell probes could also be set up as a distributed array of communication relay stations. Such a set-up was portrayed in Carl Sagan’s Contact. In this story, a dormant Bracewell probe was lying in wait in the Vega system. It began to transmit a strong signal after it received a radio signal from Earth. The device itself was part of a larger network of probes, as witnessed later by Ellie’s journey from probe to probe.

3. Worker probes

If we are going to embark on megascale engineering projects, we’re going to need robots. Lots of 'em. Projects like Dyson Spheres, Ringworlds and Alderson Disks would require fleets of specialized and artificially intelligent probes working in concert to construct these truly massive structures.

Given the sheer scale of these projects and the amount of matter that would have to be subverted, it’s not unreasonable to assume that millions of individual probes would be required. The most sensible way to construct and disseminate these probes would be through self-replication schemes.

These probes could also be put to work as mining machines that dig-out and transport matter across vast distances. Ideally, these probes would be programmed to work together and take advantage of swarming intelligence and emergent properties.

4. Colonization probes

The advent of molecular assembling nanotechnology will make it possible for probes to go about interstellar colonization. It’s conceivable that a von Neumann probe could find a suitable planet and use the matter around it to not just reproduce itself, but to establish a colony and seed actual settlers.

Such settlers would likely be uploaded consciousness patterns. This would obviously require an advanced mind emulation scheme, powerful artificial intelligence, and advanced supercomputing. Ideally, these consciousness patterns would be able to migrate to a robotic body for corporeal investigation of the environment. The number of settlers in any given location could be significant, limited only by computational resources.

Colonization probes could also construct data receivers and transmission stations so that uploaded persons could travel as digital data streams from one point to another. Consequently, the dream of traveling at the speed of light will some day be possible.

Colonization probes, sometimes referred to as seeder probes, could also perform double-duty as terraformers. Project Genesis, as portrayed in the Star Trek film series, utilized such a probe, which was able to transform a dead planet into one that suited the needs of its future inhabitants.

5. Uplift probes

Probes could also work to transform and 'uplift' other civilizations and their citizens. This scenario was explored in 2001: A Space Odyssey in which an advanced extraterrestrial civilization used probes (called monoliths) to steer the direction of evolution on Earth. In the story, these probes endowed primates with the capacity to use tools, and later, the human David Bowman was transformed into the next stage of evolution, the so-called Star Child.

This scenario was also explored in David Brin’s Uplift series in which advanced civilizations brought sapience to primitive life forms--what’s more accurately termed biological uplift. Also conceivable is technological or civilizational uplift in which an extraterrestrial intelligence brings an entire civilization up to its own advanced level.

Motivations for doing so could involve meta-ethical imperatives meant to reduce suffering, to prevent civilizations from destroying themselves, or to ensure the safe onset of non-threatening post-Singularity intelligences. Or, it could be part of your plan to take over the Galaxy.

Uplift probes could quickly bring a civilization to a post-Singularity, postbiological condition. Such a force might appear as a colonization wave that sweeps across the Galaxy, transforming all that it touches into computronium. Such a scenario has been projected by such thinkers as Hans Moravec and Ray Kurzweil.

6. Berserker probes

Unfortunately, you're going to have to look out for malevolent probes, what Fred Saberhagen dubbed Berserkers. Just as an intelligent civilization could use self-replicating probes to spread life across the Galaxy, another misguided or evil civilization could do quite the opposite and destroy everything.

Berserkers could be disseminated with the sole purpose of sterilizing every planetary system it encounters, forever eliminating the possibility for life to emerge and evolve. Should it encounter an inhabited planet, it could use any number of schemes, including nanotech instigated ecophagy, to quickly destroy all life in a matter of hours. By using a scorched galaxy policy, a civilization could sterilize the Milky Way in about 500,000 years.

Alternately, berserker probes could be disbursed across the entire Galaxy and lie dormant, patiently waiting for signs of intelligence.

Berserkers could also work to stamp out intelligent life that it deems dangerous. Anders Sandberg, Eliezer Yudkowsky and myself conceived of a strategy in which an advanced civilization (or Galactic club) could monitor for potentially dangerous post-Singularity mind-types and quickly stamp them out of existence.

7. Police probes

It’s not unreasonable to suggest that probe-making civilizations would also be thinking about defensive measures. Sandberg recently came up with an idea for anti-berserker policing probes (what I've started to call Sandberg probes). These devices would be on the lookout for bad news of any kind and take action.

Civilizations might want to establish quarantined areas; policing probes would ensure that nothing gets through the defenses and ensure the integrity of a specified region. Xenophobic civilizations might want to set up quarantined areas to prevent memetic infection, to protect themselves against invasion of any kind, or simply due to a fear of the unknown.

The best way of stopping a replicator, argues Sandberg, is to nip it in the bud. To do so, an advanced civilization would require widespread surveillance and enough power to deal with possible threats. And because replicators could emerge outside a given region of control, a civilization would want to have widely stockpiled defenses. The easiest way of doing this? Yup, you guessed it: make a replicator that spreads and builds these stockpiles and quietly waits for signs of something threatening.

So, where are all the probes?

Given all this technological potential, one must wonder why we haven’t encountered any extraterrestrial probes. Why haven't extraterrestrials communicated with us? Why haven't we be uplifted....or destroyed?

This conundrum was first articulated by Frank Tipler and has become a critical driver of the Fermi Paradox. It's been a cause of much the contact pessimism that has taken root since the 1970s (my own inclinations included). If it's so easy for probes to colonize the Galaxy, then where the heck are they? Tipler concluded that extraterrestrials simply don't exist.

Carl Sagan and William Newman came up with a different answer. They were convinced that Tipler had it all wrong and that all this talk of probes was sheer poppycock. Sagan and Newman, in their 1983 paper titled "The Solipsist Approach to Extraterrestrial Intelligence," calculated that von Neumann probes, should they exist, would eventually start to consume most of the mass in the Galaxy. Consequently, they concluded that intelligent civilizations would never dare construct such probes and would try to destroy any such device as soon as it was detected.

I'm not so convinced. Probes with even a modicum of AI and smart programming could be programmed to stop after a certain reproductive threshold has been achieved (time-to-produce schemes, maximum number of iterations, etc.). These probes wouldn't be simple mindless automatons. Moreover, the Sagan and Newman theory violates non-exclusivity; it might explain why most civilizations wouldn't dare embark on such colonization schemes, but not all. All it would take is just one.

And interestingly, Sagan and Newman seemed to be arguing for counter-measures against probes -- a strategy that Sandberg has argued would require self-replicating police probes. Moreover, as Sandberg writes,

One of the interesting things with police probes is that it makes strategic sense to announce that they are around to civilizations that might "break the law" - yet not reveal exactly how strong they are or what their modus operandi is. So the Fermi paradox appears to say that there are no police around here right now.

Further, says Sandberg, one species' police is another species' invader - we would probably not like having some alien probe impose their view of what is an unacceptable activity on us, and vice versa. And the process of making police probes will likely be indistinguishable from making other replicators. Consequently, there might be a race to set up the first interstellar police force.

At any rate, the reason for the absence of probes is still a mystery. And as the future ruler of Galaxy, you're going to have to assume this is the case. So you better get going and create a fleet of self-replicating probe before somebody else does it first.