Accelerating Future Transhumanism, AI, nanotech, the Singularity, and extinction risk.


Five Important Things from the Last Week

There's so much relevant news from the past week, I can't just focus on any one thing... so here are five of the most significant things to hit my radar in past week:

In ascending order of importance.

5. On Marginal Revolution: What are some unknown but incredibly important inventors? Why can't we get rid of the penny? And what is the moral basis of capitalism?

4. Lawrence Berkeley lab and Oxford University researchers developed a particle accelerator that takes electron beams and powers them up to a billion electron volts (1 GeV) in only 3.3 centimeters using a technology called laser wakefield acceleration. If these particle accelerators become popular and start to edge out conventional accelerators, then we'll both learn a lot more about particle physics, and put ourselves at greater risk for creating a stable strangelet. Doing a risk/benefit calculation is difficult because of uncertainty in the probabilities involved.

3. If all goes well, we may start running our automobiles on ceramic ultracapacitors which take us 500 miles on only $9 worth of electricity using a battery that recharges in 5 minutes. Eric from Digital Crusader did a few basic calculations and found that transferring that amount of energy would require a rate of 1.2 megawatts, much greater than anything seen in current home electronics systems. The inventors of this technology claim that one day it will completely replace the internal combustion engine.

2. The mouse brain has been mapped down to individual cells. It only cost $41 million to do. A 3D atlas of the common lab mouse brain can be found here. If we had computers a couple orders of magnitude more powerful than today, we could start trying to simulate that mouse's brain in a virtual environment. The success of the mouse brain mapping project is also a testimony to the success of high-level philanthropy. Paul Allen, co-founder of Microsoft, contributed $50 million to the project, more than it even needed. This Merkle paper is relevant as background.

1. Chris Phoenix proposes the creation of cubic micron DNA structures. Specifically, Chris proposed "solid molecular constructions, using DNA as a backbone, plus other arbitrary molecules precisely positioned within the volume. " He estimates that it would be possible to design one for $10 million to $100 million once the entire process is automated. The idea came out of a thought experiment about what would be possible with today's technology and only a "moderate amount of engineering".

The idea would be to build bricks that can independently manufacture other bricks, to produce a rudimentary DNA nanofactory. Less ambitiously, you could design bricks that perform specialized tasks, like breaking down garbage efficiently, and then mass-produce the bricks to perform that function. The power of the approach is that, with current technology, you can precisely specify the DNA structure within a cubic micron volume, making it possible to eventually build any structure that can be designed. Because the density of DNA is about 1.3 nm^3 per base pair, it would take about 500 million base pairs worth of DNA to fill a cubic micron space. At current DNA synthesis prices ($0.10 if it's your machine) that works out to $50 million/block, but the cost is rapidly falling.

Chris expounds a bit more on the concept here. Meanwhile, CRN argues "yes, it's coming soon".


Astronauts Collapsing, Space Hotels Launching

From CNN: No sooner does Astronaut Heidemarie Stefanyshyn-Piper return from 12 days at zero gravity on the space shuttle Atlantis and the International Space Station than she starts collapsing repeatedly at the podium while giving a speech, and has to be escorted out. This highlights what we should have known all along: humans weren't built for space. In fact, it may be impossible to grow up in a zero-g environment, because gravity is necessary for the formation of healthy bones and organs.

In Marshall Savage's book The Millenial Project, he argues that we'll all live in 3D space bubbles because it's such a better use of space than a 2D space station. In principle, he's correct - but in practice - our bodies just can't stand it, if we want to transition gracefully from space life to earth life, anyway. Therefore, it's best to build rotating space stations, like the O'Neill cylinders of yore, which unfortunately are very resource-hungry. I'm partial to the Lifeboat Foundation's Ark I design, as pictured below:

The current cost of getting a metric ton of material to orbit is $10 million, and as Ark I can reasonably be estimated to weigh about 10,000 tons, the price tag of getting all that stuff up there would be $100 billion. Not impossible, but certainly quite expensive. Obviously, we need affordable spaceflight before we can build colonies suitable for extended human stays. To lower the costs even further, atomic spaceships should be perfectly acceptable in the shorter term of solar system exploration. Using that technology, we could've had people on Mars in the 70s!


Bigelow announced at lunch that he will be putting up a three-person space station in late 2009 or early 2010, about fifty percent bigger than an ISS module. He is putting up a destination in hopes that the transportation will come along (and in order to spur the transportation providers). Station will last for several years. Will be executing contracts in 2008 for transportation contracts to Sundancer. Expects between four and eight trips (people and cargo) per year, after six-month shakedown. Then trips will commence whenever transportation becomes available. 2012 will see the launch of another module providing 500 cubic meters of habitable volume. Will support sixteen launches a year for full utilization (again, cargo and people). Minimum three-week stay, but market limited at ten million, so wants to establish private astronaut program for other nations (this is not news). Make sixty instead of eleven countries with an astronaut corps. Could represent on the order of a billion a year in revenue. Launch estimates from fifty to a hundred million per flight. About time to take human spaceflight from the exclusive domain of governments. Will be changing that in the next half decade.

Also, NASA recently reshuffled its budget, eliminating many side programs in favor of a central focus on the lunar program. Many space enthusiasts have intensely complained, but honestly - I don't mind a Lunar Focus, and neither do many scientists. It's all fun and good to study asteroids, use inferometry to look for exosolar planets, poke around Europa, and the like, but does it actually help get our butts off this rock in the next few years? Not really, and that should be our main priority - putting our eggs in more than one basket.

By the way, could you subscribe to this feed if you've been reading this blog via RSS? It would be convenient in terms of keeping everybody on the same page, and the URLs for other feeds may soon be removed.

Filed under: space 45 Comments

The Box

Filed under: risks 7 Comments

SL Callgirl Interested in the Singularity

In a display of knowledge that eclipses that displayed by many transhumanists, Khannea Suntzu, a "high-priced and highly-desired call girl" in SecondLife, speaks with Wagner James Au of New World Notes, the most widely read blog on SecondLife. Khannea is apparently fairly well-known, even outside of SL, as has been interviewed or profiled on the popular gaming portal 1 Up, and even the prestigious French publication Le Monde. When she contacted Wagner James Au, it was not to give an interview (which she's done many times before), but to merely use him as a loudspeaker to talk about the Singularity:

"What does interest me is of an entirely different nature," she continued. "The fact is, I have a bit obsessed with the idea of a singularity. You may have heard of it. Not everybody believes such a thing may actually happen, but after having read Ray Kurzweil's The Singularity is Near, I am deeply concerned. In his last two books he talks about this whole singularity business and sketches a prospect that it'll be actually possible that in the next 10-20 years some part of what we are could be captured in some form of artificial medium. Maybe 'we' could be imprinted in a new substrate, effectively copying or transplanting 'us' into a new state of existence.

Not sure that we'll choose to upload our entire bodies and minds by 2020, even if it were possible, but she continues:

"However, in the next 10-20 years there's gonna be a whole lot of things happening. Starting with better VR and robotics, in the 2020s we'll have computers that can think, hard AI, nanotechology, genetic therapies and maybe the first affordable life extension therapies. Most people around, maybe even you, have no clue what is gonna happen. By 2020 nearly all low-education jobs in the modern industrial societies can and will be replaced or streamlined by automated systems. You'll see unemployment numbers in Europe, US and Japan ranging in the 20-30%. By 2025 production costs of all objects you can buy in the stores will be dropping fast because of nanofactory replication. That will create even more unemployment. By 2030 we'll have actual mind-machine interfaces of some sort. What happens after that, I can't even begin to speculate. It could all very well lead up to a real singularity, with some kind Artificial Intelligence (whether or not it IS self-aware) improving itself in spectacular increments. By 2050 we may have a world completely alien to the world we live in right now. I don't know for sure but I expect spectacular things."

What is she going to do about it?

"These ideas of mine effectively make me a bit of an obsessive," she allowed, "though hopefully not Cassandra. I can go and circlejerk endlessly in the transhuman scene, but I've had enough of that. My primary interest is starting to instill some sort of public awareness, preferably inside SL.

The transhumanist scene (not "transhuman") is certainly not a circle jerk, because new people are getting involved all the time, and the subjects for discussion are so broad that people exchange new knowledge regularly. I highly doubt that Khannea has even joined a transhumanist mailing list, otherwise she would at least have the "transhumanist" word down, but I'm sure she's probably skimmed the transhumanism entry on Wikipedia.

So instead of participating in a circle jerk in "transhuman scene", she's going to be instilling public awareness in SL... perhaps during some of her call sessions? I don't know. I always thought it was better to instill public awareness on a website that anyone can find in seconds via Google.

"I'll be in my 60s in 2030. I have a fair chance of being alive by then, if the world doesnt suffer thermonuclear meltdown before then. If technology indeed does the unprecedented jumping through hoops I think it can, I want in on all that. And my guess is the best way to achieve that is to leave a mark, influence people, create awareness, foment discussion and maybe even political activism. I don't want my already small chances to experience all this to be halved into the single percent digits. I am convinced that what I do and say, right now, can increase the chances, with whatever modest resources I have, of me seeing some spectacular sights in the next decades - or maybe the next centuries.

Influence people to do what? Be excited that technology will inevitably give us nanofactories, AI, and life extension therapies without any effort on our part? This is "passive singularitarianism", and it's quite silly.

"And yes," she concluded, referring to the brunette avatar in fetish wear, "Khannea fits a central role in that. In fact, it is my sincere ambition to be her. I think that I have a fair shot of experiencing her, being her, to the best of my ability, before 2040. Maybe even well beyond 2100."

Good luck! Now that you're excited about The Singularity is Near, you should definitely take a shot at reading a substantial amount of other transhumanist writing.

In the end, it is my position that the web and real life will give us much greater leverage than SecondLife. If you have something to say, put it in HTML form. If you want to secure large donations for transhumanist projects, work out deals on the phone or in person, because hitting someone up on SL probably won't take you very far. The comments on the interview show the general cluelessness of the SL poplace when it comes to transhumanism. Perhaps log out of SecondLife and use Google for a while?

Also, Anne shares her thoughts on Artificial Intelligence and the Singularity, which she's just starting to get acquainted with.

Filed under: singularity 43 Comments

Lifeboat Foundation on

The Lifeboat Foundation has been mentioned on a recent article on

Scientists could generate a black hole as often as every second when the world's most powerful particle accelerator comes online in 2007.

This potential "black hole factory" has raised fears that a stray black hole could devour our planet whole. The Lifeboat Foundation, a nonprofit organization devoted to safeguarding humanity from what it considers threats to our existence, has stated that artificial black holes could "threaten all life on Earth" and so it proposes to set up "self-sustaining colonies elsewhere."

But the chance of planetary annihilation by this means "is totally miniscule," experimental physicist Greg Landsberg at Brown University in Providence, R.I., told LiveScience.

The point of this is not whether or not there really is an immediate risk from nuclear accelerators (which is debatable), but that an organization focused exclusively on existential risk is getting this kind of coverage from a top-tier science website.

Existential risk is a pretty big deal. To quote Derek Parfit's Reasons and Persons (1984):

I believe that if we destroy mankind, as we now can, this outcome will be much worse than most people think. Compare three outcomes:

1. Peace
2. A nuclear war that kills 99% of the world's existing population
3. A nuclear war that kills 100%

2 would be worse than 1, and 3 would be worse than 2. Which is the greater of these two differences? Most people believe that the greater difference is between 1 and 2. I believe that the difference between 2 and 3 is very much greater... The Earth will remain habitable for at least another billion years. Civilization began only a few thousand years ago. If we do not destroy mankind, these thousand years may be only a tiny fraction of the whole of civilized human history. The difference between 2 and 3 may thus be the difference between this tiny fraction and all of the rest of this history. If we compare this possible history to a day, what has occurred so far is only a fraction of a second.

Humanity's future potential matters, a heck of lot more than whether or not you personally make it through this century. If you want to be part of a community devoted to preventing existential risk, join us.

Filed under: risks 14 Comments

AGI Workshop in Palo Alto

Here are my messy notes from Sunday... and here are the powerpoints.

Here at the Crown Plaza in Palo Alto for the Artificial General Intelligence Research Institute's AGI Workshop... people are just starting to show up, and it's almost 2:00PM. The audience appears to be a balanced mix of AI researchers and general transhumanists/futurists, with very substantial overlap, of course. The first presenter is Ari Heljakka, a programmer with Novamente LLC and the CEO of Finnish AGI company GenMind Ltd. His talk is entitled "AGI & the Singularity". It's a general overview of Singularity ideas from Vernor Vinge, Eliezer Yudkowsky, Ben Goertzel, and Ray Kurwzweil... Eliezer walks in while a slide of his is being shown, he's wearing the same shirt as in the slide, everybody laughs.

AGI vs. Narrow AGI is the second segment, it a short talk by Ben. Look at the powerpoint if you're interested.

Third segment is a panel with myself, Christine Peterson, Ben Goertzel, and Eliezer. Christine seems to be unclear on the power of a superintelligent AI, thinking in terms of police rather than utility fog in every cubic centimeter of the earth's crust. Stay tuned for the release of this video, it's interesting.

Fourth segment:

Ben Goertzel, PhD
Novamente LLC
Biomind LLC
Artificial General Intelligence Research institute
Virginia Tech, Applied Research Lab for National and Homeland Security

Biggest commenters: Jef Allbright, Brad Templeton, Eliezer Yudkowsky.

Ben Goertzel again. Topic: Novamente, a Practical Architecture for Artificial General Intelligence. This focuses in on the AGI part, and temporarily puts aside the Friendly AI part. Goes over AI estimates - mentions that the estimates have come down a lot. Used to be hundreds or thousands of years, now it's 50-100, even with academic AI researchers. Ben mentions that he's more optimistic than Kurzweil, who says 2029, and says that if it takes that long, it'll be due to sociopolitical reasons, not technical reasons. Mentions that predictions are tied to outcomes - self-fulfilling prophecies. Early 80s - time travel, unification of physics, AI, origin/synthesis of life. Thought that his impact in AI would be greatest - "all it does it figure out the right code to write".

Brad Templeton: decide to work on time travel, if your future self doesn't come back, it's definitely intractable.

Long-term goal: AGI. Approach based on computer science algorithms, though human psychology is utilized. Working on a couple of narrow AI projects, reusing code from AGI. Scalable C++ run on linux machines. Biomind OnDemand for bioinformatic data anlysis, ImmPort:NIH Web biomind/novamente backend.

Patternist philosophy of mind: Recognizing patterns in world and itself. Probability theory used to quantify and relate patterns. Logic (term, predicate, combinatory), used as a base-level language for expressing patterns. Reflectivity of recognizing self-patterns and improving those patterns. Philosophical approach matters for how you approach it technically. Carrying out procedure P in context C will achieve goal G.

Ben books: Probabilistic Term Logic, Engineering General Intelligence, Artificial Cognitive Development. Open source: AGISim, embodiment for AI. Goal-driven inference, 45%, sensor processing 35%, background inference, 20%. Ben is not a purist about embodiment, cares about the ingestion of databases (WordNet, FrameNet, Cyc, quantitative scientific data, etc.); approach called 'post-embodied AI'. Non-body AI is okay, but it would have a hard time getting together with humans. Eliezer: AI is made of math, so even if it seems easy, it's actually that hard to do Friendly AI all the time.

Formal stage: Reasoning about situations different than anything you've ever experienced? Reflexive stage: deep understanding and control of self structures and dynamics. Full self-modification: something humans can't do. Ben lists various types of databases which could be integrated at various stages. Cyc, Mizar database, etc.

As long as it rewrites the schedulers and knowledge representation, it's still reflexive, when it rewrites the C++ itself, then you're in a new domain. Eliezer: Capability is more important than whether or not it actually is called a different thing. Atoms = nodes or links. Atoms have truth values (probability + weight of evidence) and attention values (short and long term importance). Atomspace weighted, labeled hypergraph. Steve Omohundro: do you keep track of the source? Ben; in many cases you can't, but you can still represent in terms of nodes or links. Jef Allbright: what about context? Ben: most links are context links. In any finite system: 'alienation of knowledge' - it can't always be put directly in context. Atomspace is not a neural net, nor a semantic net. Markov networks, Bayes networks - this is not one of those either. What processes are taking place? This is cyclic rather than acyclic. Accurate Bayes - too computationally intensive, heuristics always required for approximation. Meant to approximate Bayesian judgements.

Low short-term importance and long term = useless. Low short, high-long = remembered but not used right away (mother's phone #), high short, low long = used then forgotten, like precepts. High long, high short = used and remembered. Truth values. Weight of eveidence low, strength low = weakly suspected to be false. Strength high, weight low = weakly suspected to be true.

Mind agents implement cognitive processes, atom space is passive. Gigabytes of knowledge in the machine - can be likened to a giant blackboard. Simplified workflow - precepts, active memory, feelings and goals, execution management, active schema pool, the world, and back again. Attention allocation - "simulated economy" approach. Pattern mining of the AtomTable, embody as predicates. SystemActivityTable, which MindAgents did what to which Atoms at which times. Nodes which fire together, wire together, language generation, probabilistic chaining, importance propagation. ABC, deduction, induction, abduction.

Filed under: AI, singularity 9 Comments

Major Study on Public Nano-Awareness

From Yahoo news:

WASHINGTON, Sept. 18 /PRNewswire/ -- Research findings released today from the first major national poll on nanotechnology in more than two years indicate that while more Americans are now aware of the emerging science, the majority of the public still has heard little to nothing about it. The poll also finds that the public looks to the federal government and independent parties to oversee nanotechnology research and development. These results, according to experts, necessitate increased education and stronger oversight as a means to increase public confidence in nanotechnology.

The poll, a telephone survey of 1,014 U.S. adults, was commissioned by the Project on Emerging Nanotechnologies at the Woodrow Wilson International Center for Scholars and conducted by independent research firm Peter D. Hart Research Associates in August 2006.

Findings reveal that one in 10 Americans have heard a lot about nanotechnology and 20% say they have heard some -- nearly double the number of Americans aware of the technology in 2004. But, 42% of Americans have no awareness of it at all. Older Americans and women, the groups most likely to use consumer products containing nanotechnology materials such as skin-care products and cosmetics, are the least informed about nanotechnology. Importantly, those individuals with an awareness of nanotechnology are more likely to believe the benefits of this emerging technology outweigh the potential risks -- supporting experts' assertions that open discourse and information about nanotechnology is crucial in establishing positive sentiment among the American public.

"Nanotechnology is increasingly being incorporated into Americans' daily lives, from the sunscreens that they wear to the golf clubs they swing and the computers they operate. With this increased use, specifically in those products that are ingested or used topically on the skin, it is essential that the public be aware of what nanotechnology is and the benefits and potential risks it presents," said David Rejeski, director, Project on Emerging Nanotechnologies. "Our purpose in conducting this poll is rooted in the very heart of the Project's mission: to spark discussion about nanotechnology's promise and about potential health and environmental risks. The findings indicate the time is ripe for government and industry to do more to educate and engage Americans about nanotechnology's tremendous potential and to conduct strategic research into possible risks."

Findings from the survey also indicate that majorities of Americans feel that the federal government (55%), universities and independent researchers (54%) have a role in overseeing scientific and technological advancements such as nanotechnology rather than relying on self-regulation by private companies and industry, especially for non-prescription over-the-counter (OTC) cosmetics and skin care products. In fact, only 12% of consumers believe the companies that manufacture these products should be exclusively responsible for regulating their safety.

Findings from the Public Poll on Nano Awareness and Trust include:

  • Men aged 18 to 49, adults with at least a college education, and adults with higher incomes are more likely to have heard about nanotechnology.
  • Fifty-nine percent of adults age 65 and over have not heard about nanotechnology.
  • More than one third (35%) of the public believes that the risks will outweigh benefits, 15% think the benefits will outweigh risks and 7% say that the risks and benefits will be about equal. Forty-three percent are not sure.
  • More than half (52%) of women are not sure about the benefit/risks trade-off, compared with 34% of men. Fifty-eight percent of older women (age 50 and above) are not clear about the trade-off.
  • Americans who have heard a lot about nanotechnology and men aged 18 to 49 are most likely to believe that the benefits of nano outweigh the risks. Almost half (46%) of those who have heard a lot, and 25% of men aged 18 to 19 think benefits will outweigh risks.
  • Political party affiliation is not a factor affecting opinions on oversight responsibility.

Pretty interesting results. If we were to do the same poll with molecular manufacturing (MNT), a subset of nanotechnology, the numbers would likely be far lower. By the way, my general position on nanotechnology is that the risks outweigh the benefits, though most transhumanists would disagree with me. Whether or not the risks or benefits or greater, the technology will be developed regardless.


Peter Thiel Gives Methuselah Foundation $3.5M

From the Methuselah Foundation website:

Peter Thiel puts his weight behind Dr. Aubrey de Grey's engineering blueprint for alleviating the debilities caused by aging

San Francisco-- Peter A. Thiel, co-founder and former CEO of online payments system PayPal, Founder and Managing Member of Clarium Capital Management, a San Francisco-based hedge fund, and angel investor in social networking site Facebook, has announced his pledge of $3.5 Million to support scientific research into the alleviation and eventual reversal of the debilities caused by aging, to be conducted under the auspices of the Methuselah Foundation, a charity co-founded and Chaired by Dr. Aubrey de Grey.

Mr. Thiel commented "Rapid advances in biological science foretell of a treasure trove of discoveries this century, including dramatically improved health and longevity for all. I’m backing Dr. de Grey, because I believe that his revolutionary approach to aging research will accelerate this process, allowing many people alive today to enjoy radically longer and healthier lives for themselves and their loved ones.

Mr. Thiel will donate a total of $500,000 over the next three years to fund pilot research projects intended to deliver early stage validation of the "SENS" approach to combating the debilitation caused by aging.

Additionally, from now until the end of 2009, Mr. Thiel promises to match every Dollar donated to the Methuselah Foundation for SENS research with a 50 cent matching contribution from himself, up to a maximum of $3 Million of matching funds.

Dr. de Grey said "I am extremely grateful to Peter for his bold and visionary initiative. I have been working with leading biologists and biochemists around the world in identifying promising research projects, and with this generous donation we will go to work straightaway."

About SENS

SENS (Strategies for Engineered Negligible Senescence) is a detailed plan for alleviating the debilitation caused by human aging. SENS is an engineering project, reflecting the fact that aging is a medical condition and that medicine is an engineering discipline. Aging is a set of progressive changes in body composition, at the molecular and cellular level, which are side-effects of essential metabolic processes; each of these changes has the potential to be mitigated and eventually reversed. Further details of SENS can be found at:

I was fortunate to meet with Aubrey when he was in San Francisco to formally accept Thiel's gift, talking with reporters and filming a promotional video. Aubrey attended a meeting of BA-Trans that I organized. Other attendees were Bruce Klein and Susan Fonseca-Klein, who were original founders of the Immortality Institute, and Adam Kamil, a frequent poster on ImmInst who flew in from LA to be at this event. Some photos can be downloaded here. Following is a photo of myself and Aubrey after a few drinks:

Filed under: transhumanism 10 Comments

First-Stage Nanoproducts and Nanoweaponry

By 2020, and potentially as early as 2010, we will know enough about carbon chemistry, kinematic self-replication, and nanoscale positional control to build a desktop nanofactory - a machine that uses many trillions of tiny arms to put together macro-scale products. Because tiny arms can move incredibly fast, they will be radically productive. It has been estimated that a 100 kg nanofactory will be able to manufacture its own weight in product in about three hours, perhaps less.

Nanofactory technology will begin with an assembler - a reprogrammable molecular machine capable of making a copy of itself. An assembler would be extremely small, composed of maybe a couple million atoms. This is about the same as a ribosome. For a reference, see this picture of some nanoparts next to a virus:

An assembler would basically be an artificial ribosome. Ribosomes are the little machines in the cell that manufacture every protein in your body. Its basic design hasn't changed in over a billion years.

Feasibility arguments for molecular nanotechnology (MNT) are well-documented in the literature. Its not a question of if, but when. The technological and sociological impact of personal nanofactories (PNs) is certain to be extreme. If regulations permit it, you will be able to construct, right in your very home, just about any structure allowed by the laws of chemistry and available feedstock. All current manufacturing, communication, and transportation processes will be fundamentally restructured over a period of mere years or even months. The first nanofactories are likely to use carbon feedstock, meaning most of the products will be made out of diamond. Water may be used as a ballast for some diamond products.

Products built using MNT will be extremely cheap: around the cost of their raw materials. This is because human labor, the primary cost of manufacturing today, is largely subtracted from the equation. Carbon is extremely cheap, and can be mined by the megaton from practically anywhere. Power requirements are modest. Made of diamond, a nanofactory will not require much maintenance.

Quickly, typical products made of plastic, ceramic, or metal will be redesigned to accommodate the new diamondoid medium. There will be diamond plates, diamond tables, diamond cutlery, ovens, coffee makers, microwaves, tiles, walls, chairs, televisions, cameras, printers, scanners, shelving, windows, computers, pens, notepads, pottery, showerheads, and so on. Something like 90% of all manufactured products will be replaced by diamondoid versions. This is what Neal Stephenson was thinking when he wrote a book called The Diamond Age.

The father of nanotechnology, Eric Drexler, lists a few things which would become possible with MNT on his website:

desktop computers with a billion processors
inexpensive, efficient solar energy systems
medical devices able to destroy pathogens and repair tissues
materials 100 times stronger than steel
superior military systems
additional molecular manufacturing systems

MNT has been called "magic", and the word choice is not entirely inappropriate. We will be able to build products with greater performance and more diverse functionality than anything you or any university Ph.Ds have imagined. All shortages of energy, food, water, and shelter will be rapidly solved, as long as nanofactories are made available to developing countries. Subdermal heaters, nanoproducts designed to do little more than generate waste heat, will eliminate the problem of obesity practically overnight. The size and range of products will be limited only by whatever regulations are built into the first round of nanofactories. And I hope that these regulations are extremely strict. You see, nanofactories will be the most dangerous technology that mankind has ever faced, thousands of times more dangerous than nuclear weapons.

Given an unrestricted nanofactory and a few million dollars worth of programming and engineering, here are a few products that I could manufacture in almost arbitrary quantities, given a couple months manufacturing time:

sniper rifles that weigh less than 5 kg, capable of firing a lethal projectile at Mach 10 towards any target within my line of sight.
extremely light and strong armor capable of stopping 10 kg explosive shells moving at faster than 10 km/sec.
Metal Storm systems which fire as many as 1,000,000 projectiles per minute through ballistics arrays.
UAV swarms capable of actively neutralizing very large rockets, providing comprehensive area denial, working together to disassemble buildings, etc.
highly maneuverable VTOL craft able to destroy almost any number of F-22 Raptors or F-35 Lightnings.
gigawatt-class, solar array or nuclear-powered microwave beams capable of completely melting tanks, aircraft, destroyers, incoming missiles, etc. from hundreds of miles away.
isotope separation systems that enrich uranium efficiently, at great speeds, giving enough fissile material to make bombs in days rather than years.
gigantic lenses capable of redirecting sunlight towards arbitrary coordinates in extremely high concentrations; a solar furnace.
missile swarms composed of individual missiles about 1 meter long, carrying 1 kg warheads, manufactured by the millions, capable of traveling through the upper atmosphere and surviving reentry.

Because products made out of diamond can be extremely strong and light, 100 kg of carbon gives you a very large bang for your buck. For example, a Mercedes S-class today weighs about 2,000 kg, but with diamondoid building materials, this weight could be reduced tremendously, if desired - the primary motivation to preserve the vehicle's current weight would be the preservation of inertia, rather than engineering limitations. An automobile made out of nanodiamond could have an absurdly low weight, on the order of a hundreth of an ounce, not including fuel. If this sounds fantastic to you, take a look at what is already possible today:

This tiny block of transparent aerogel is supporting a brick weighing 2.5 kg. The aerogel's density is 0.1 g/cm^3.

Anyway, the point of all this is simple: nanofactories need to be extremely restricted in the products they can build, or there is going to be big problems. The open source, anti-digital rights management, P2P-generation needs to get this. Information may want to be free, but if weapons designs are readily available and manufacturable in the post-MNT world, there are going to be problems of the likes we've never seen. To minimize the risk of danger, the safest option is to have all product designs authenticated by a central authority. Yes, that scary phrase, "central authority". This central authority needs to be capable of determining which designs are safe, maintaining an extremely high level of nanofactory security, and enforcing the law when people try to circumvent it. The libertarian dream of minimalist government, unfortunately, must be discarded.

Now, in general, I'm extremely against big government. It can be a huge waste, and extremely inefficient relative to market-driven competition. But when it comes to managing magic, decentralized solutions simply won't do. There needs to be a global standard and global regulations. Rogue states won't do, either. One rogue nation could use MNT to manufacture enough weapons to turn the capitals of any opposing nation, no matter how large, into a series of smoking craters. This is a risk we shouldn't be willing to take, and once the potential of MNT starts to sink in with higher-level government officials, they won't.

Life extensionists: realize that the greatest risk to living longer is not actually aging, which we will eventually defeat cleanly, but existential risks of the type I frequently discuss, including superintelligence and nanotech arms races. You can extend your expected future life more by lowering the probability of these disasters than through any other means.


Better Ways to Get into Space

When it comes to getting into space, traditional rocketry is the pits. Gigantic tanks that cost millions of dollars, massive fuel requirements, trajectories that fight against the atmosphere instead of using it to their advantage. Out of the five space shuttles built, two have gone boom. If you're going to build a Lifeboat in orbit, deploy solar power satellites, or visit space hotels, you're going to need a better way to get into space.

Luckily, there are numerous ideas, including rocket planes, orbital airships, the space elevator, and the space pier. 3/4 of these ideas already have companies putting serious resources towards their realization. Let's take a look at the details, shall we?

The rocket plane is currently the idea getting the most attention and funding. All current rocket planes are only capable of taking people to the edge of space and back, rather than going into orbit. Thus, trips on rocket planes are called suborbital space flights. The typical rocket plane launch consists of a larger plane that helps a smaller unit ascend to about 14km, where the air is several times thinner than at ground level, then releasing it.

The first rocket plane to reach space, as defined by 100km altitude, was the North American X-15, which was flown almost two hundred times throughout the 60s. Today, we have SpaceshipOne, with SpaceShipTwo on the way by 2008. Virgin Galactic, the company largely funding the present effort, has stated that if SpaceShipTwo is successful, it will follow up with a craft capable of making it into orbit, SpaceShipThree:

In general, craft that suck material directly from the atmosphere to use for oxidation offer superior specific impulse to traditional chemical rockets:

Earlier this year, Virgin Galactic started constructing the world's first purpose-built commerical spaceport, SpacePortAmerica, in southwest New Mexico. Space Adventures Ltd. is partnering with the people behind the Ansari X Prize to plan and eventually build the Ras Al Khaimah spaceport in the United Arab Emirates, and the Singapore Spaceport in Singapore. Here's what they would look like upon completion:

Spiffy, yes?

So what is an orbital airship? Proposed by JP Aerospace, the orbital airship concept is a three-staged process which includes a conventional airship, a permanent sky base, followed by a helium-filled, solar-powered ascender unit that slowly accelerates horizontally until it reaches escape velocity. Here's a look into the crotch of the Y-shaped ascender unit:

Here's a look at the inside:

The JP Aerospace website has several videos of both real flights and CG mockups. Problem is, engineers on the newsgroup confirmed that their plan is physically impossible. You can't gather enough energy with solar cells to overcome the atmospheric drag on the ascender unit. Interesting idea, but seems as if it will require major breakthroughs to be feasible, if ever. Time to move on to the space elevator...

The space elevator is a concept being championed by the Liftport Group. It's one of the older alternative space ideas, dating back several decades. The proposed contruction method, as I understand it, is to guide an asteroid into geostationary orbit, launch a series of rockets filled with carbon nanotube fiber to it, and lower a thin "seed elevator" to the earth's surface. From this point on, the elevator could be strengthened by using robotic climbers to add additional material to the initial thread.

The space elevator concept is very popular, in no small part due to its common presence in sci-fi, Tower of Babel-esque connotations, and the numerous CG mockups floating around on the net. I used to be a big space elevator advocate, but I've started to think that in the short term, it is not the ideal means of getting to space. To quote extropy list veteran Eugen Leitl:

I have problems with terrestrial space elevators (much less so with lunar elevators), largely because of need of actively moving the ribbon to avoid perforation by debris, because the tensile strenth required is borderline to what physics gives you, with not much safety margin, and if you fail only once you've wrapped all your infrastructure around the equator.

Forming a continuous rigid strand from earth to space, a space elevator would interfere with all sorts of low earth orbits, and also be an ideal target for terrorists of the future. Once that cable snaps, it would practically take the power of a god to grab the two ends and reconnect them without imminent disaster. My preferred alternative is the Space Pier:

The Space Pier is an idea from J. Storrs Hall, a pioneer in the field of nanotechnology and my colleague on the CRN Global Task Force. His site explaining the idea can be found here. Essentially, it's a 100km-tall, 300km-structure topped with an electromagnetic linear accelerator. Air resistance at this altitude is lessened by a factor of one million, and the plan is less cumbersome and catastrophe-prone than a taut string that reaches six earth radii from the surface. It would be a compressive tower, that is, standing under its own weight rather than using a geosyncronous counterweight. If one of the legs were K.O.ed by a nuclear terrorist attack, the structure as a whole would still stand. Utility fog distributed around the legs would provide yet another failsafe. The trip to the top is much shorter than climbing up a 36,000 km space elevator, and the way to low earth orbit is fast and easy. Once in LEO, one could employ ion drives or other techniques to get to GEO or out of the earth's gravity well.

Well, that's my summary... dozens of other exotic earth-to-orbit schemes can be found on this site. Happy flying. ;)

Filed under: space 31 Comments

Scientists Abandon AI Project After Seeing The Matrix

A classic article from The Onion.

Filed under: friendly ai No Comments

Keith Elis on the Financial Future of AGI

Keith Elis' statement from a discussion on the newly-created Singularity list:

Hi all,

There is a possibility that at some point in the future, government
agencies, wealthy foundations, and non-profits will significantly
increase expenditures as to AGI development. When the purse-strings
open, and the money flows, it will flow like tax dollars, bequests, and
donations do -- toward politically tenable projects. Yudkowsky's
Friendliness theory, whether you agree with it's technical feasibility
or not, is very effectively positioning the Singularity Institute's
future AGI projects to be Politically Friendly.

In the summer of 2003, the US media reported on an attempt by DARPA to
put a futures market in place which would ostensibly be able to forecast
certain undesirable events such as terrorist attacks, assassinations and
the like. The idea was to find a way to elevate our awareness before a
threat materialized, and so DARPA was studying these prediction methods.
Now, while the idea itself has theoretical and practical merit, some
members of Congress sensed an easy victory against an injured opponent
and piled on. There was no debate; there was no thoughtful consideration
of the project's chances for success; there was no collective desire to
learn more about the idea. The project was incinerated by fiery
political soundbites, and no opposing voice was willing to be
incinerated along with it. The incident caused Adm John Poindexter to
resign in disgrace and we will never hear of the US government playing
with prediction markets again in our lifetimes.

There is a lesson here for everyone working on an AGI project that "just
needs funding to get there." You must be politically tenable. Funding
your project must be justifiable in a soundbite. And it should take a
Ph.D. droning on and on for pages in technical jargon to present an
argument against you, your theory, your design, and your most likely
outcome. As an exercise, and remembering that you're really, really
smart, and the rest of us aren't, how do you debate against the
following statement?

"We should ensure, in fact guarantee, that AGI doesn't wipe out

Do you not see that lining up against this statement for whatever
technical mumbo-jumbo reason is suicidal? But forget funding for a
moment. Think about what happens when Congress gets involved in
regulating this field, and guys in jackboots come knocking. Is it
smarter to have publicly stated 'Friendly AI is bunk' or to have said
'It's the only kind AI worth building'?

Please, everyone working on a real AGI project that might hasten a
Singularity, you must learn this lesson from Yudkowsky. He's not wrong.


The point of this message is that explicitly including a well-thought-out Friendliness strategy in an AGI project is likely to garner more approval from the higher-ups. Not to mention minimizing the risk of the human race being replaced by indifferent machine intelligence.

Filed under: friendly ai 8 Comments