Accelerating Future

From New World Notes, James Au’s standard blog on SL, comes news that Philip Rosedale, founder of Linden Lab (the company behind SecondLife), is “working towards creating a sentient artificial intelligence which [exists] in a virtual world.” Um, OK. Here is a quote:

That Philip plans to revolutionize AI technology — in effect, achieving singularity in a virtual world — isn’t that surprising, because he said as much when I talked with him for The Making of Second Life:

Hey! Slow down. A Singularity in a virtual world could easily transcend its boundaries and cause problems for us all. I don’t think that Rosedale will achieve human-level AGI anytime soon, but we must acknowledge that the boundary between the virtual and the “real” is pretty damn thin. Perhaps this is just James Au’s choice of words, but I wish all of his readers would understand that AGI is not just a game, or a new entertainment company. It could utterly transform the planet in a way that it’s never been transformed in its entire 4.56 billion year history.

8 Responses »

Casually visiting CNN.com this morning, I was rather surprised to see a prominent ad including a picture of my friend Wendell Wallach, co-author of Moral Machines: Teaching Robots Right from Wrong and the Moral Machines blog. Here is the ad (click it to see it in context on CNN’s site):

Amazing. CNN gets about five million daily visitors, and I’m sure Honda is advertising their short video in many other venues, including possibly television. Here is the video:

I have mixed feelings about this pro-robot corporate messaging. My hope is that the need for Friendly AI will become more clear as we develop more advanced robots and AIs and observe their propensity to follow the letter of the rules we give them while ignoring the “spirit”, and we’ll realize that the cognitive capabilities needed to interpret the spirit of requests are above and beyond what is needed to achieve goals in the world. While robots that mindlessly follow the letter of requests aren’t that big a deal when robots are weak and dealing in narrow tasks, it will become progressively more annoying and apparent when robots become more capable and perform across a range of tasks.

The next decade will surely bring many exciting demos in robotics and AI that I very much hope will bring more people around to our position that a well-funded Friendly AI research program is necessary if Homo sapiens is interesting in surviving the 21st century. Maybe Honda could fund one. Here are some demos that I could gain attention:

1. A “robotic engineer” that uses autonomous robots built in fab labs to extend its physical infrastructure. At the recent Foresight conference, Hod Lipson stated that a major goal of his lab at Cornell was to design a robot that can autonomously move out of the 3D printer as soon as it is fabricated. (Cool, huh?) Beyond that would be the goal of using such robots to build additional 3D printers automatically, without human intervention. I predict that both of these milestones will be met by around 2022.

2. An autonomous program in SecondLife that overloads servers by generating a large quantity of self-replicating objects, like the Sonic rings that caused 600,000 users to experience serious lag in late 2006. When I used to go on SecondLife, two of my favorite toys were 1) an object that allowed me to break the usual 100 m (or whatever) limit on flight height (you can fly by default in SecondLife), 2) an automated construction program that allowed me to generate hundreds of objects in a couple seconds. My hobby was flying up at about 600 m, constructing giant hollow spheres with many hundreds of tiles, then letting them rain down on random parcels below. Not very nice, but pretty funny. I was receiving automated messages scolding me for cyber-littering for weeks.

What would really be interesting is an automated script that spoofs SecondLife accounts, then uses them to create thousands of independent scripts that each generate a large quantity of garbage objects. I’ve never heard of anyone doing this, but I do know that various “hacker” groups on SL in the past, especially the Patriotic Nigras, have aggressively pushed the griefing envelope. (The motto of Patriotic Nigras is “ruining SecondLife since 2006″.) Their custom SecondLife client, ThugLyfe, includes various scripts and ban-evasion techniques that create major headaches for well-meaning Lindens. It would be interesting to create a griefing-oriented AI-based SecondLife agent and see if it could independently reinvent any of the functions included in ThugLyfe. This environment could be a proving ground for any would-be AGI program.

The reason why all the above is important is that virtual worlds will continue to increase in complexity until they better and better approximate the real world, so any AGI that would be a risk in the real world will hopefully be tested and used in virtual worlds, where we can hear about them. Philip Rosedale also supposedly started an AGI company called LoveMachine, but I don’t have much of a clue about it.

3. Some demo involving adaptive control of microid swarms. Microids are interesting because this is one of the first designs of an ant-sized microbot that avoids using complex moving parts that are susceptible to wear from friction or dust jamming. I’d be interested to see if the recent spray-on glass technology could be used to better seal ant-sized robots with complex moving parts against dust. 3D printers fabricating microids is a longer-term vision, but maybe by 2022? I am somewhat skeptical that useful microscale-fabrication desktop fabs are possible — my intuition is that we’ll only see macroscale-fabrication desktop fabs and maybe nanofactories (but hopefully not) before the Singularity.

4. A robotic swarm that powers itself from a diverse set of energy sources, including biomass, solar, and wind. This would probably be possible now if anyone put the money towards it.

Anyone else have any other ideas?

2 Responses »

New staff bios have been added to Singularity Institute’s team page: short bios for Research Fellows Anna Salamon and Steve Rayhawk, Media Director Michael Anissimov, and Chief Compliance Officer Amy Willey. Check them out, and feel free to ask if you’re interested in knowing more about what each staff member does. Here’s the lineup:

Cross-posted from SIAI blog.

2 Responses »

No Responses »

At Less Wrong, Roko Mijic claims that despite survey results, most philosophers are not really “strong” moral realists, and in fact their “non-realist” moral stance is often anti-realist for all practical purposes. Perhaps unsurprisingly, I totally agree.

No Responses »

Here’s a extended abstract presented at the 2009 Asia-Pacific Conference on Computing and Philosophy that is making the rounds. The point of the paper, which was written by Carl Shulman, Nick Tarleton, and Henrik Jonsson, is that consequentialism as commonly discussed has a number of “free variables” where intuitions disagree about the right values of these variables. Therefore, machine ethics should draw on the emerging field of moral psychology to figure out how to fill in these free variables. This point is plainly put in the title of one of the last sections, “Current moral theories are inadequate for machine ethics”.

A reply from UK philosopher David Pearce has recently been posted by Roko Mijic at Less Wrong. Roko points out that David is a moral realist, i.e., that he believes there is a fact of the matter about what is right and wrong.

My immediate response to David would be that I agree with him that superhappiness is possible and worth pursuing, but to say that the propositional content of goals does not matter for humans, AIs, or any sentient beings is going way too far. A world guided by happiness as the sole optimization criterion will eventually toss away moral complexity to become quadrillions-of-smiling-faces land. The world ought to be shaped by a complex optimization criterion, and David doesn’t really address this. I can only speculate that maybe he finds it hard to believe that sentient beings would discard complex propositional content in their morals. If a magical force preserves that complexity, that would be nice, but I prefer the approach where it is deliberately protected.

6 Responses »

Here’s an interesting article if you haven’t seen it yet. Did you know that the Marie Byrd Land is the largest patch of unclaimed land on Earth, with an area of 1.6 million sq km?

No Responses »

6 Responses »

From yesterday’s list of links, I particularly want to call attention to the rotifer link. This press release is interesting because it shows how animals can survive even when they are exact genetic copies of one another. Instead of outcompeting parasites through mutation, they run away by going into cryptobiosis. I predict that a form of asexual multicellular synthetic life will be created by 2030 that can defend against parasites through aggressive defense, say silica spines, so that running away isn’t even necessary. These organisms will just sit around and reproduce. The primary method to get rid of them at first will be dessication, but this will eventually prove useless as they disperse too widely to target.

What many humans don’t realize is that we are surrounded by quintillions of organisms with very little genetic diversity that dominate us in terms of biomass and persistence. They are the status quo — we are the aberration. These are organisms that have survived every mass extinction. Culprits include the tardigrades (which can survive outer space), nematodes (absolutely ubiquitous; it is estimated there are between 10¹⁸ (one quintillion) and 10²¹ (one sextillion) nematodes worldwide, and they are crawling all over you right now), chaetognaths (considered useful models of basal bilaterans, there are a lot of them in the oceans, really a lot), and so on.

The only reason that these organisms aren’t ripping us all to shreds right now is because there have been no synthetic biologists to push them out of evolutionary minima and give them more sensible strategies for total domination. Sorry to be alarmist, but I studied evolutionary biology for a couple years and that is my opinion. Evolution is terribly poor at transversing local minima to reach a global optima, and that is really the only saving grace for fragile macroscale multicellular agglomerations like ourselves. Interesting and low-energy-cost evolutionary innovations are rarely combined because they require several working parts to come together which are maladaptive individually but adaptive in cooperation.

The reason why rotifers are interesting is that their lack of genetic diversity makes them a good model for self-replicating machines. The ability to switch into a dormant, armored state (cryptobiosis) seems characteristic of a variety of small organisms, and we can expect this ability to be exploited to the fullest by human-engineered microscale replicators. The ability to distribute many of these replicators across a wide area will eventually create a “viral load” scenario analogous to the one faced by aging humans — so many diverse beings build up in our body that the workload faced the immune system to combat nascent infections eventually becomes prohibitive and the system breaks down.

Some scientists have laughed at the idea that human-engineered organisms could dominate microbes that have evolved for billions of years, but I find this ridiculous. Human-engineered artifacts have already outperformed everything created by evolution in terms of energy density, speed, mass, acceleration, local dominance, and so on. The key point is that evolution is radically dumb (but it has many trials available) and humans are very smart. Let’s discuss some of the ways to engineer microorganisms that cannot be defeated by the legacy biota.

1. Broad-spectrum biocides: natural organisms use a variety of biocides, but observe that humans have created thousands of highly effective synthetic antibiotics and biocides that evolution never discovered even after four billion years of experimentation.

2. Phage-immune bacteria, for instance bacteria that use genetic programs incompatible with malicious code injection by phages. Phages are the main bacteria-curtailing force on the planet and we depend on them for our survival.

3. Bacteria specifically engineered for immunity to broad-spectrum antibiotics which produce and secrete these antibiotics as a biofilm. There is even the possibility of release-and-shield, where microbes release the biocide then shield themselves from it for long enough for the competitors to be defeated, at which point the shield is raised.

4. Sucking them in: microorganisms could coat themselves in a gel shield which absorbs and dissolves both nutrients, phages, and rival microbes. For instance, the extracellular matrix of animal tissues is much stronger than the slime layer used by bacteria. Cooperative colonial bacteria could create stronger extracellular shields depending on how well-established the colonial region is, devoting stronger shields to the colonial center and weaker shields to the exploratory fringes.

5. Incubation-then-release: many evolutionary minima involve colonial organisms that are evolutionarily strong in larger colonies but evolutionarily weak in small colonies. By sterilizing a large area, filling it with nutrients, and allowing a founder population to develop (a “mega petri dish”), an important evolutionary minima could be hopped.

6. Quorum computing: evolution has developed a variety of means for microbes to communicate with one another on a crude level: quorum sensing. One of the interesting evolutionary innovations of the last billion years was to produce multicellular organisms that survive against many uncooperative microbes. By creating microbial superorganisms that effectively cooperate and compute using biocomputation, it may be possible to beat multicellular life at its own game by creating “organisms” miles across that effectively cooperate to defeat all rivals. This is definitely not a near-term risk but it could be a risk within the lifetimes of many alive today, given no singleton that guards us at a low level.

7. The last point in particular opens up a very large space for experimentation. For a colony that knows how to differentiate its perimeter members from interior members, it can activate all sorts of interesting genes in the perimeter members to make life miserable for organisms next to them. Bacteria already do this in a rudimentary way with quorum sensing. As long as a suitable barrier can be erected, the production of a variety of poisons is possible and safe for the majority of the colony.

Even natural selection in hospitals is enough to create killer bacteria immune to many antibiotics. What about bacteria specifically engineered by smart humans for reproduction and survival?

4 Responses »

Neuroscientists making computers smart enough to see connections between brain’s neurons
At Davos, MIT faculty discuss the nature of intelligence
Rotifers avoid sex for millions of years by blowing away
Insectlike ‘microids’ might walk, run, work in colonies
The Physical Basis of High-Throughput Atomically Precise Manufacturing
Chris Phoenix: Basic Survival Package
Air Force to use artificial intelligence and other advanced data processing to hit the enemy where it hurts
Robin Hanson: AI In Far And Near View
The Growing Backlash Against Overparenting
MIT’s Mind-Machine Project: Student Position Paper
Law of probabilities backs hopes for E.T., conference hears (what a banal claim.)
NEC’s Facial Recognition Technology Achieves First Place in the Still-Face Dataset
Cube Spawn — Open Source, Flexible Manufacturing System
Mondo 1995: Up and Down With the Next Millennium’s First Magazine
Dale Carrico: Futurological Brickbats
The Onion: God’s Wrath According to Pat Robinson

Thanks to everyone who sent me links in the last week.

One Response »

My friend Miron Cuperman, a software entrepreneur here in San Francisco, recently launched an interesting information security initiative called Gitian. The motivation is to eliminate software distributors as a single point-of-failure for malicious code injection. Here is Miron’s blog post which summarizes the initiative:

Operation Aurora (Google’s compromise by China) highlights the possibility that software distributions may be targeted for code injection by malicious parties. If Apple, Microsoft or Linux distributors are compromised, a large percentage of individuals, businesses and governments could be consequentially compromised when they install software updates.

One way to mitigate such a risk is to have multiple independent security auditors sign software distributions. This is more likely to be successful in an open-source environment, where source is available and can easily be inspected. I started such an initiative in late 2009 – Gitian.org.

When Miron told me about the initiative, I asked him to outline some specific failure scenarios to better illustrate what the initiative is meant to protect against, which he did here. The Gitian site is here. In his blog post, Miron also points out that a deterministic build system is necessary to implement the security measures he is promoting, and that before he implemented Gitian, he had never run across one, even though it is relatively straightforward to create.

No Responses »

From RepRap blog:

The Foresight Institute has announced its Kartik M. Gada Humanitarian Innovation Prize to design and build a better RepRap. There is an interim prize of $20,000, and a grand prize of $80,000. They consulted with the core RepRap team before the announcement and we were initially concerned that the prizes might drive developers to secrecy in order to give themselves a competitive edge. As you will see they have addressed those concerns by making it a condition of winning the prize that solutions should be pre-published and made available under a free licence. For ourselves and on your behalf, we would like to thank the Institute for the enthusiasm that these prizes demonstrate for the RepRap project and for their magnificent generosity.

Congrats to Foresight Institute and Kartik Gada for establishing this interesting and substantial prize. There is another prize, too. Besides the Personal Manufacturing Prize, there is a Water Liberation Prize, described here:

The winner of the Water Liberation Prize of up to $50,000 will be the first person to invent a device that is either solar powered, manually cranked, or otherwise not dependent on the existence of an electrical grid, can produce at least 4 liters of potable (drinkable) water per day, either condensed from the air (as measured in approximate 50% ambient humidity) or filtered through a nanomembrane, and can be mass-produced (as demonstrated by a pilot run of no less than 100 units) for a cost of less than $5 per unit. The filter should be washable and re-usable, without requiring a periodic supply of new filters, as the device may be used in areas without access to a suitable distribution channel.

8 Responses »