Transcendence

The Disinfornation Encyclopedia of Transhumanism and the Singularity

Red Wheel/Weiser, LLC

Contents

ABOLITIONISM

(See also Designer Babies)

This may be crazy, but what say we end suffering in all sentient beings? David Pearce, an English transhumanist, proposes abolitionism: a new movement to completely end all such sentient misery. And you thought Europeans weren't ambitious anymore.

The primary methodology would be germline engineering of "designer babies" for maximum happiness and minimum pain, in combination with various other advanced technologies for ending scarcity and eliminating physical vulnerabilities in humans.

PEACE AMONG THE ANIMALS

But it doesn't stop there. Pearce, an animal rights enthusiast, figures, Hey, let's make the animals peaceful and happy! "Gene therapy will be targeted both on somatic cells and, with even greater forethought, the germline. If cunningly applied, a combination of the cellular enlargement of the mesolimbic dopamine system, selectively enhanced metabolic function of key intra-cellular sub-types of opioidergic and serotonergic pathways, and the disablement of several countervailing inhibitory feedback processes will put in place the biomolecular architecture for a major transition in human evolution." Quite a mouthful, but if you're trying to end all suffering on Earth, it figures to be a bit complicated.

James Kent: Pearce's intellectual embrace of paradise engineering places him on the cusp of a modern philosophical movement that eschews Darwinian fatalism and looks to a post-Darwinian future where humans are freed from the cynical bonds of genetic expression and natural selection.

David Pearce: Here are some grounds for cautious optimism that transhumanism can eliminate suffering in humans:

1) We Shall Soon Be Able to Choose Our Own Level of Pain Sensitivity

A revolution in reproductive medicine is imminent. Clearly, our emotional response to raw pain is modulated by the products of other genes. But recent research suggests that variants of the SCN9A gene [see "Pain's in the Genes" Science Magazine, March 8, 2010, or take Pearce's word for it] hold the master key. Thus, in a decade or two, preimplantation diagnosis should allow responsible prospective parents to choose which of the SCN9A alleles they want for their future children—leading in turn to severe selection pressure against the SCN9A gene's nastier variants.

At present, we can't envisage safely choosing one of the (extremely rare) nonsense mutations of SCN9A that eliminates physical pain altogether. A future world of nociception without any phenomenal pain at all will depend on advances in neuroprosthetics and artificial intelligence. Yet by selecting benign alleles of SCN9A both for ourselves and our children, the burden of suffering can soon be dramatically diminished.

2) We Can Soon Choose How Rewarding We Want Our Daily Life to Be

The brain is a dauntingly complex organ. Yet the biological roots of mood and emotion are primitive and neurologically ancient. Their metabolic pathways are strongly conserved in the vertebrate line and beyond. A research paper published in Nature in 2007 illustrates how the presence or absence of a single allele may dramatically enrich or impair the quality of one's entire life.

How can we guard against unanticipated side effects from novel gene therapies? What will be the inevitable "unintended consequences" of life-changing innovation? And what will be the societal implications of a population biologically predisposed to enjoy richer and happier lives?

Genetic tweaking to promote richer experience is just a foretaste of posthuman sentience. I predict that our descendants will enjoy gradients of genetically preprogrammed bliss every day of their lives.

ARTIFICIAL GENERAL INTELLIGENCE

(See also Artificial Intelligence, The Singularity)

The original goal of the AI field was the construction of "thinking machines"—that is, computer systems with human-like general intelligence. Due to the difficulty of this task, for the last few decades the majority of AI researchers have focused on what has been called "narrow AI"—the production of AI systems displaying intelligence regarding specific, highly constrained tasks. Common narrow AI applications include the AI linguistics underlying Google and other search engines; the AI planning and scheduling software used throughout the military and industry; the AI fraud detection software underlying modern credit card operations; the AI used by those friendly folks at the NSA to find patterns in your phone calls; and the AI gaming software underlying everything from IBM's chess-playing Deep Blue to the bots in massively multiplayer online games. These are all amazing achievements (albeit the uses can be questionable) that have a common narrowness of scope, which is why Ray Kurzweil has characterized them as "narrow AI."

In recent years, however, more and more researchers have recognized the necessity and feasibility of returning to the original goals of the field. Increasingly, there is a call for a transition back to confronting the more difficult issues of "human level intelligence" and more broadly "artificial general intelligence (AGI)."

AGI describes research that aims to create machines capable of general intelligent action. The term was introduced in 2003 in order to avoid the perception that the field was only about creating human-level or human-like intelligences, which is also covered by the term "Strong AI." AGI allows for the inclusion of nonhuman as well as human models of general intelligence.

One approach to Artificial General Intelligence may involve reverse engineering the brain. This quest is described in detail in Jeff Hawkins' book On Intelligence.

Written with Surfdaddy Orca

When Will We Get Human-Level AIs?

Ben Goertzel, Seth Baum, Ted Goertzel: When will we have Artificial General Intelligences (AGIs) we can talk to? Ones as smart as we are, or smarter?

The majority of the experts who participated in a 2010 AGI survey by h+ were optimistic about AGI coming fairly quickly, although a few were more pessimistic about the timing. However, all the experts in the study, even the most pessimistic ones, gave at least a 10 percent chance of some AGI milestones being achieved within a few decades. The experts were asked to give estimates on each of four milestones:

• Passing the Turing test by carrying on a conversation well enough to pass as a human.

• Solving problems as well as a third grade elementary school student.

• Performing Nobel-quality scientific work.

• Going beyond the human level to superhuman intelligence.

There was consensus that the superhuman milestone would be achieved either last or at the same time as other milestones. However, there was significant divergence regarding the order of the other three milestones. One expert argued that the Nobel milestone would be easier than the Turing Test milestone precisely because it is more sophisticated: to pass the Turing test, an AI must "skillfully hide such mental superiorities." Another argued that an AI that passes the Turing test needs the same types of intelligence as a Nobel AI "but additionally needs to fake a lot of human idiosyncrasies (irrationality, imperfection, emotions)" Finally, one expert noted that the third grade AI might come first because passing a third grade exam might be achieved "by advances in natural language processing, without actually creating an AI as intelligent as a third-grade child." This diversity of views on milestone order suggests a rich, multidimensional understanding of intelligence. It may be that a range of milestone orderings is possible, depending on how AI development proceeds.

One observed that "making an AGI capable of doing powerful and creative thinking is probably easier than making one that imitates the many, complex behaviors of a human mind—many of which would actually be hindrances when it comes to creating Nobel-quality science." He observed, "humans tend to have minds that bore easily, wander away from a given mental task, and that care about things such as sexual attraction, all of which would probably impede scientific ability, rather than promote it." To successfully emulate a human, a computer might have to disguise many of its abilities, masquerading as being less intelligent—in certain ways—than it actually was. There is no compelling reason to spend time and money developing this capacity in a computer.

ARTIFICIAL HIPPOCAMPUS

(See also Implants, Prosthetics, Neurotechnology)

In the part of our brain called the hippocampus, our short-term memories become long-term memories through a process involving the movement of electrical impulses through neurons. A number of projects have looked for ways that an injured—or maybe just insufficient—hippocampus can be bypassed via the implantation of miniaturized electronic equipment.

In a set of experiments announced in 2013, neuroscientists from the University of Southern California (USC), Wake Forest University (WFU), the University of Kentucky, and Defense Advanced Research Projects Agency (DARPA) have been able to use a device to record patterns in tissues of mice and monkeys related to specific memories, and then replicate those by stimulating healthy brain cells with electricity—making up for functions lost to damage in other brain cells.

Researchers are predicting that humans could be using such devices to replace lost memories in five to ten years. The FDA may be willing to hurry this along, given that the immediate application will be for Alzheimer's, an ever-growing crisis that will likely be even more critical by that point.

Projecting a little bit further out, artificial hippocampus devices are a giant step toward the transhumanist dream of devices that can record and preserve all our memories.

David Pescovitz: Biomedical engineer Theodore Berger at the University of Southern California in Los Angeles has developed an artificial hippocampus. To do this, Berger built mathematical models of neuronal activity in a rat's hippocampus and then designed circuits that mimic those activities. Joel Davis at the Office of Naval Research, a sponsor of Berger's work, said, "Using implantables to enhance competency is down the road. It's just a matter of time." While Berger's work is a far cry from a hard drive for the brain, I'm intrigued by the notion of being able to "back up" my memory just in case.

ARTIFICIAL INTELLIGENCE

(See also Neurobotics, Robotics)

As opposed to AGI (Artificial General Intelligence), which doesn't really exist yet, Artificial Intelligence (instruments that can autonomously perform specific constrained tasks) is all around us. In addition to search engines, credit card fraud detection, snooping on phone calls, and gaming, AIs are answering our phones and giving customer assistance (badly!), beating humans at chess and Jeopardy, driving cars, targeting drone strikes, doing most of the flying that we attribute to our airline pilots, and collecting data and doing rudimentary analyses of conditions in space. It's even correcting things I type here, although sometimes I wish it would stop.

As we all know from dealing with automated phone systems, one of the limitations of constrained AI is that real life situations often don't correspond to a fixed set of options. The distinction between current AI and AGI (or humans or even animals) is its minimal ability to respond to unexpected stimuli.

Transhumanists are most enthused about developments in contemporary AI that appear outwardly to mimic human abilities like learning and showing emotions.

Steven Kotler: UC Irvine neuroscientist Jeff Krichmar's AI-based bots develop personalities because, instead of preprogramming behaviors, these robots have neuromodulatory systems or value judgment systems (move towards something good, move away from something bad) that are modeled around the human's dopaminergic system (for wanting, or reward-based behaviors) and the noradrenergic system (for vigilance and surprise). When something salient occurs—in the case of Krichmar's bots, that's usually bumping into a sensor in a maze—a signal is sent to its brain telling the bot to react to the event and remember the context for later. This is conditional learning and it mimics what occurs in real brains.

ARTIFICIAL LIFE

(For artificial biological life, see Synthetic Biology)

EVOLUTION IN DIGITAL FORM

How can something as complex as life arise without a Creator? How can mere humans even hope to replicate it? Well, it turns out that it's not as difficult as you might think. For many years artificial life (or ALife) researchers have been simulating biological life using computers (or with robotics or biochemistry). Right now, you could be playing God in the simplified world of a browser window.

DOTS ON A GRID

Stanislaw Ulam and John von Neumann first discussed cellular automata in the 1940s, but the field really took off when John Conway unveiled his elegant Game of Life in 1970, and there are now many free versions online. They all use a simple grid and simple rules. Each square either contains a dot ("alive") or is empty ("dead"). A dot stays alive if it has two or three neighbors, or dies if it's too crowded (four or more neighbors) or too lonely (one or no neighbors). When an empty square borders three dots, a new dot is "born" there. Now, start with a bunch of dots and watch the fun.

LET THERE BE ALIFE

Surprising patterns will emerge. Still lives (such as a square block of four dots) will stay static. Oscillators (the simplest is three dots in a row) will cycle between two or more states. Gliders and other spaceships will fly across the grid. Glider guns will emit streams of gliders. Puffer trains will crawl, leaving persistent smoke. Collisions will create more patterns. Moving, eating, excreting, birth, death—all visible in dots on your screen.

OPENWORM

The open source OpenWorm project has produced the most advanced digital life form so far. The nematode Caenorhabditis elegans, with all of its one thousand cells and the neurons connecting them, is modeled entirely in software. The latest version wriggles realistically and at the same speed as a real one, though this is due to hard-coded instructions, not the 302-neuron brain, which is still being developed.

SO WHY DOES ALIFE MATTER TO TRANSHUMANISTS?

Several reasons. The Game of Life is "Turing complete": it can simulate any real-world computer or computer language. It's used in cryptography. As a model it's helpful for anyone researching synthetic biology. Some physicists even believe that our universe is best described as a cellular automaton.

But perhaps the core lesson is that the complex (and solutions to complex problems) need not be designed from the top down: astounding levels of complexity can arise from simple rules plus randomness. Remember that the next time someone claims that big solutions require big bureaucracies, or when someone scoffs at the empowering possibilities of transhumanist tech.

AUBREY DE GREY

(See also Longevity/Immortality, The Methuselarity, SENS Research Foundation)

Somewhere around the turn of the millennium, Aubrey de Grey, an English biogerontologist who is now as famous for his Father Time beard as he is for his outspoken vision of radical life extension, looked at aging as an engineering problem and decided: Eureka! We can do this.

Since then, de Grey has become the preeminent spokesperson for the scientific quest to end death. He formed the Methuselah Foundation, which offers a cash prize for a particular longevity breakthrough, and the SENS (Strategies for Engineered Negligible Senescence) Research Foundation, which is conducting scientific research on longevity, using principles de Grey laid out in his 2007 book, Ending Aging. He has appeared on 60 Minutes, The Colbert Report, and the Barbara Walters special report "Live to 150, Can You Do It?" and recently wrote a guest opinion piece as part of a 2013 Time magazine cover story on Google's recently announced project to conquer death.

THE REAL DE GREY

De Grey is an easygoing fellow known for his fondness for beer and for generally not adhering to the intense physical regimens followed by many other longevity fanatics. Having ridden in the back seat of his car one time, I can report that de Grey likes depressing British post-punk, which totally humanizes him as far as I'm concerned.

Aubrey's Approach

Michael Anissimov: Instead of exclusively studying the complex biochemical processes of aging in detail, as in gerontology, or ameliorating the worst symptoms of age-related decline, as in geriatrics, de Grey and his supporters advocate an "engineering approach" to aging that asks, what are the main categories of age-related biochemical damage, and how can we fix them? The idea is not to eliminate the sources of agerelated damage, but to fix the damage fast enough so it doesn't accumulate and cause health problems. This is far easier than deciphering all the intricacies of the biochemistry of aging ...

It's Not Personal

Aubrey de Grey (interviewed by R.U. Sirius): I'm actually not mainly driven by a desire to live a long time. I accept that when I'm even a hundred years old, let alone older, I may have less enthusiasm for life than I have today. Therefore, what drives me is to put myself (with luck) and others (lots and lots of others) in a position to make that choice, rather than having the choice progressively ripped away from me or them by declining health. Whether the choice to live longer is actually made is not the point for me.