May 24th, 2006
|mathemajician||02:41 pm - Hard AI Future Salon with Yudkowsky|
For those of you with fast internet connections, here is a two hour talk given by Eliezer Yudkowsky on the safety of super intelligent machines and related topics.
I felt that in some parts of this talk he dwelt too much on things that aren't very important, and some of the questions from the audience made no sense at all, however I do think that Yudkowsky did have some good points to make.
May 22nd, 2006
|spoonless||10:12 pm - James Hughes lecture|
For those in the Bay Area, Dr. James Hughes--Executive Director of the World Transhumanist Association--is giving a free public lecture at UC Santa Cruz this Thursday (May 25th). If you picked up one of our flyers at the Singularity Summit (I know a lot of people did), this is a reminder to make it down to Santa Cruz. Hope to see you there!
What does it mean to be human, and are there limits on what humanity can or
should become? Is it ethical to design our children with genetics and
cybernetics? Will we some day want to give intelligent cyborgs and robots the
same civil rights afforded to humans? Under what circumstances (if any) is
human cloning and the mixing of human and animal DNA ethical? Should it be
legal to alter your state of mind using psychoactive drugs? Can we improve
intelligence, mood, and attention span using psychopharmacology, and if so is
it ethical to do so? Is death a natural and inevitable part of life, or will
people continue to live longer and longer as medical understanding of cellular
aging processes improves, perhaps one day leading to a cure for aging
entirely? These are the types of the issues Dr. James Hughes will explore this
Thursday at UCSC.
( Read MoreCollapse )
April 23rd, 2006
|bobstark||01:55 pm - Playing God|
Throughout history, biologicalevolution has produced many solutions to technical problems ofsurvival. When the lion needed food, it was given teeth, figurativelyspeaking. When the bird needed to flee preditors, it had wings. Whenthe humanoid needed to do many of these things and did not have any ofthese features, they had tools to manipulate their environment.
We don't need wings because we can build planes. We don't need teeth because we have guns. And so on.
The tools that most allow us this power of environment manipulation are:
1. Hands, fingers (physical manipulation)
2. Reasoning, memory (cognitive manipulation)
3. Language (cooperation)
Weare really the species that controls its own evolution, in this sense.How exciting! The way this evolution occurs is, then, throughtechnology.
How does AI fit into the grand scheme of evolutions?
Howwill it help us evolve ourselves? Well... there are nanomachines withintelligent features that could do just about anything in the future while inside our bodies,there are "agent" technologies to do tasks for us to let us focus onother things in our lives, and then there is creating robots tocompletely replace humans in some areas.
Insofar as this means simulating manyhuman characteristics, this is called "strong AI" - it's what you seein the movies. This is to be constrasted with the form of AI that I'mgenerally interested in - something like utilitarian/weak/whatever AI -the injecting of intelligence into everyday items to complete tasksthat otherwise couldn't have been done, but not creating robots to talkto or anything.
Playing gods is the next step in our evolution,in the sense that we can create things to do tasks for us (not so muchin the organism-creating sense). This is AI.
Is that megalomaniacal? Maybe. So sue me.
April 21st, 2006
|mathemajician||11:30 am - Machine invention|
While the contents of this article are not all that new, it's still a nice example of how machines are starting to outperform humans in areas such as design and invention:
Koza asks Jones to pull up the stats on the wide-field telescopic eyepiece. Amid a rush of figures, he reads off the number “295.” That’s how many generations it took for genetic programming to engineer around the Koizumi-Watanabe patent. In fact, the invention machine’s lens is better than the Koizumi-Watanabe system: Because it keeps breeding until all design specs are met, often some performance requirements are exceeded by the end of the run. The final field-of-view for Koza’s eyepiece is a remarkable 10 degrees higher than the 55 degrees achieved by Koizumi and Watanabe.
Of course this is just a taste of things to come. In 15 years computers will be 1,000 times as powerful as they are now, and no doubt the theory of how to use this raw computation power to discover new designs and inventions will have significantly improved also. The shift from human intelligence to machine intelligence is underway...
In other news, upcoming hard disks from Seagate will have capacities up to 750 GB. We're almost at 1 Tera Byte hard drives now. To put that in perspective, the number of bytes on such a drive is ten times greater than the number of neurons in a human brain.
April 6th, 2006
Rough cut of a feature film about AI, robots, the singularity, and the 21st century, titled “Building Gods,” featuring Nick Bostrom, Kevin Warwick, Hugo de Garis and Anne Foerst. From Four Door Films, 1 hr 20 min. Watch it here.
March 30th, 2006
|mathemajician||11:05 am - First commercial 'spaceline'|
Richard Branson's Virgin Galactic is to start of offering sub-orbital trips into space starting in 2008. A ticket will costs $200,000. Reservations can be made today.
March 23rd, 2006
|mathemajician||03:51 pm - Will Japanese Robots Rule the World by 2020?|
Every time I go to one of the large international AI conferences there is yet another Japanese robotics demonstration with new tricks to show off. Clearly robotic technology is moving very fast, but is it so fast that Japanese robots might Rule the World by 2020?
I doubt it, but I do predict that by 2020 we will be thinking about such scenarios a lot more seriously than we are now.
March 15th, 2006
|mathemajician||11:09 pm - The end?|
As I'm sure most of you remember, a month ago singularity_now temporarily turned into a messy battle between those who view the singularity as being about science and technology, and those who view it in mystical and spiritual terms. I wanted a singularity community focused on the former view and I figured that if singularity_now wasn't going to provide this, then I should go and start a community that did.
For the first week after I created the community here, singularity_now continued to attract off topic and mystical posts to the list until finally the moderator stepped in and started removing them. Since then, singularity_now has been on a pretty strict science and technology diet. I'm not sure if it's because the moderator has been enforcing this, or because the mystic crowd decided to post elsewhere. In any case, the need for sci_singularity is now in question.
On one hand sci_singularity is growing in popularity by about ten people a week. So given time it could become a reasonably popular community. On the other hand, if singularity_now continues to stick to science and technology, then having another community doesn't make much sense.
I'm happy to go either way. If people want the community here to continue and want to post to it, fine with me. And if people think it's time to pack up and move on then that's fine with me too --- after all I'm plenty busy enough actually trying to make the singularity happen! Thus I'll turn it over to the community to decide... of course if you don't care either way, then just don't register a vote.
End the community
Continue the community
UPDATE: Ok, after a couple of days the poll results are 9 votes (including one by comment) to 3 in favour of continuing the community. Thus the community will go on.
March 8th, 2006
|mathemajician||10:21 am - The Singularity Investor|
If the coming changes in technology are going to turn the economy upside down, were should people put their money in order to be on the winning side of this shift? This is something I've thought a bit about myself, I figure that investing in companies that own fabrication plants could be good. My reasoning is that once powerful general intelligence algorithms and architectures have been discovered, the demand for CPUs and RAM will go through the roof. Somebody else who has been thinking about similar things is the Singularity Investor.
Speaking of such things, Adaptive AI (a.k.a. a2i2) today announced that they are looking for more employees following a successful funding round. Moreover, in part of the announcement they have stated that they "are on schedule to achieve robust human-level learning and cognition within 2 years." Interesting. [Disclosure: I hold equity in this company, but I haven't been involved with their AI development work for about 3 years.]
And in case you missed it, Intel have announced that their new Conroe desktop processor should increase performance by 40% and decrease power by 40%. Their upcoming server CPU called Woodcrest will increase performance by 80% and decrease power by 35%. The first results coming out of the benchmarkers at AnandTech seem to support this. Yep, the exponential growth in computer power is marching along very nicely.