The Coming Artilect War
June 3rd, 2010Via: Technofascism Blog:
If you haven’t heard of Hugo de Garis, AI researcher and Professor at China’s Xiamen University, then you’ve probably never heard of an Artilect either. A contraction of Artificial Intellect, the Artilects are future creations that, according to de Garis, will be immortal, godlike robotic beings that, due to being built on a substrate of nanoscale computer chips, will have billions of times the intelligence of the average human. Needless to say, these Artilects would not for a second hesitate in wiping out the entire human species if their need dictates; much like a human wouldn’t trouble themselves over running over an ant colony on their way to work.
It’s important to listen to de Garis because he is one of those rare scientific researchers that actually considers the implications of his work. In de Garis’ view, in the late 21st century humanity will have the technological ability to create the Artilects. At that point, humans will be divided into three camps: the Terrans, who will attempt to violently prevent the creation of the Artilects; the Cosmists, who think building them is a good idea; and the Cyborgists, who will attempt to merge with the Artilects to preserve some shred of their humanity. These three camps will fight it out and the winner will decide the fate of the planet. Since the war will be fought with late 21st century weaponry, the death toll is expected to be in the billions.
In a Forbes article, Hugo de Garis describes this coming Artilect war:
I differ sharply with well-known futurist Ray Kurzweil on his over-optimistic prediction that the rise of the artilect this century will be a positive development for humanity. I think it will be a catastrophe. I see a war coming, the “Artilect War,” not between the artilects and human beings, as in the movie Terminator, but between the Terrans, Cosmists and Cyborgists. This will be the worst, most passionate war that humanity has ever known, because the stakes–the survival of our species–have never been so high. Given the period in which this war will occur, the late 21st century, with late 21st century weapons, the scale of the killing will not be in the millions, as in the 20th century (the bloodiest in history, with 200-300 million people killed in wars, purges, holocausts and genocides) but in the billions. There will be gigadeath.
Although many people might dismiss de Garis’ vision as science fiction, as a fellow ex-AI researcher myself, I believe that de Garis’ horrifying prophecy is fundamentally correct.
Personally, I’m not too worried about the AI-apocalypse in my lifetime. I do worry about autonomous killing machines though. You don’t need some transcendental intelligence to rig a robotic shotgun with motion and heat sensors and drop it in a neighbourhood of “militants”.
The reasons I’m not particularly worried about AI go something like this:
– For the AI-apocalypse scenario, the AI has to be mobile and miniaturised.
– The current architecture of computing hardware is unsuitable for emulating brains. That could change, but I still think it’s a fair way off. For true AI, I think we’re going to need something intrinsically parallel and analog rather than serial and digital.
– Energy requirements. Intelligent life has had millions of years to evolve efficient ways of powering brains. Unless we also go the organic route (ie: making our own biological intelligence), then the “Artilects” are going to need a source of electrical energy that does not compromise the mobility and size requirements.
(I’ve just finished re-reading Society of the Mind http://www.amazon.com/exec/obidos/ASIN/0060176946/ref=nosim/cryptogoncom-20 which touches on this sort of thing. Worth a read.)
I guess it’s mainly about optimizing existing practices though.
“True AI” ? It has to come with objectives. Even if its objective is to find goals. Otherwise : how would it assign priorities ?
My friend suspects this : pure AI wouldn’t do $hit.
True irony?
The development of genuine AI capability will only be realized after the fact, in observation after initial system power up. In other words, if such a technology does open any sort of Pandora’s Box, it will probably be too late already.
Of course, as many reading here already know, a host of different sci-fi & future tense authors have already deeply mined this subject, leaving me with little else to add.
Meanwhile, I too expect to see “autonomous” robotic armaments of US manufacture firing live in Eurasia before 2020. They’ll not be genuinely autonomous, first-generation units being more akin to fly-by-wire drones than anything AI.
And it’s the stuff they can field now that bugs me most anyway. Imagine what significant numbers of land-based Phalanx-type ballistic weaponry positioned in strategically hardened positions would accomplish in terms of “bringing home the troops”. A major shift in military history there, akin to the stirrup enabling mounted cavalry.
Mix in the latest generation of every-shrinking power supplies and various esoteric weaponry in DARPA’s darker corners over the next ten years or so — and the rush to develop new space-based delivery platforms — and no AI whatsoever is required for a human dystopia writ large.
Hell on Earth.
The only strategy left appears to be positioning yourself as the lowest value target as is possible under the circumstances. Easier said than done in the era of the modern surveillance state slowly shrinking inward around the digital domain.
When officially-sanctioned state/corporate regulated online “credentials” are the systemic norm, identity fraud and theft will become even more prevalent, when it is the only route left to individuals interested in establishing some sort of genuine anonymity. Probably already is, come to think of it.