Please register to participate in our discussions with 2 million other members - it's free and quick! Some forums can only be seen by registered members. After you create your account, you'll be able to customize options and access all our 15,000 new posts/day with fewer ads.
I searched for the term “singularity” in this thread but did not find it. Forgive me if this has been explained already but the thread is over 1000 posts.
From Wikipedia:
The technological singularity (also, simply, the singularity)[1] is the hypothesis that the invention of artificial superintelligence will abruptly trigger runaway technological growth, resulting in unfathomable changes to human civilization.[2] According to this hypothesis, an upgradable intelligent agent (such as a computer running software-based artificial general intelligence) would enter a "runaway reaction" of self-improvement cycles, with each new and more intelligent generation appearing more and more rapidly, causing an intelligence explosion and resulting in a powerful superintelligence that would, qualitatively, far surpass all human intelligence. Stanislaw Ulam reports a discussion with John von Neumann "centered on the accelerating progress of technology and changes in the mode of human life, which gives the appearance of approaching some essential singularity in the history of the race beyond which human affairs, as we know them, could not continue".[3] Subsequent authors have echoed this viewpoint.[2][4] I. J. Good's "intelligence explosion" model predicts that a future superintelligence will trigger a singularity.[5] Emeritus professor of computer science at San Diego State University and science fiction author Vernor Vinge said in his 1993 essay The Coming Technological Singularity that this would signal the end of the human era, as the new superintelligence would continue to upgrade itself and would advance technologically at an incomprehensible rate.[5]
At the 2012 Singularity Summit, Stuart Armstrong did a study of artificial general intelligence (AGI) predictions by experts and found a wide range of predicted dates, with a median value of 2040.
Ridiculous? I have no idea. But that number, 2040, catches one’s eye.
And now you're way off into theoretical/philosophical land, which has nothing to do with AI replacing most workers. It's like saying that because one of Toyota's welding machines, which can take three or four complex parts, put them into position while allowing for manufacturing tolerances, then weld them better than a human can in a tenth of the time... isn't a robot because it doesn't look like Robbie (or Data).
When do you think that AI would be able to examine several machined parts, realize that the assembly would have less friction and better robustness if it were designed differently... then go through those designs, being mindful of cost, of customer-acceptance, of attitudes of the company executives... coordinate with the subject-matter experts in aerodynamics, in NVH, in acoustics, in drivetrain-development and so forth... run said tests and computations, and finally, weld together a prototype... which it would then demonstrate on the track... and give a convincing presentation to management?
I searched for the term “singularity” in this thread but did not find it. Forgive me if this has been explained already but the thread is over 1000 posts.
I don't think it's been mentioned. I would be the most likely contributor to mention it, but I was avoiding the complexities of the topic.
Quote:
At the 2012 Singularity Summit, Stuart Armstrong did a study of artificial general intelligence (AGI) predictions by experts and found a wide range of predicted dates, with a median value of 2040.[/i]
Ridiculous? I have no idea. But that number, 2040, catches one’s eye.
That's for a very specific case, something like achieving sustained fusion. AGI is the usual term for human-level or human-surpassing machine intelligence. It's a decade or so off at best, by most opinions. (Oh, and 2012 is medieval history at this point in the AI field. )
But the next tier of jobs to fall to "robotics" or "automation" or "AI" or "veeblefetzerization" does not need that level of breakthrough. AI won't replace doctors in the next five years, but it might invade hospital wards for routine patient monitoring and communication. It won't replace architects and front-line engineers, but it will probably replace routine CAD work and guided engineering development. It won't replace CEOs, but it will almost certainly replace most of HR, accounting, finance and customer service. And in tech... it's going to replace whole layers of employment, guided by a few designer- and architect-level humanoids.
None of those need to be a machine personality you'd marry, be best friends with or call in Captain Kirk to destroy with circular logic.
When do you think that AI would be able to examine several machined parts, realize that the assembly would have less friction and better robustness if it were designed differently... then go through those designs, being mindful of cost, of customer-acceptance, of attitudes of the company executives... coordinate with the subject-matter experts in aerodynamics, in NVH, in acoustics, in drivetrain-development and so forth... run said tests and computations, and finally, weld together a prototype... which it would then demonstrate on the track... and give a convincing presentation to management?
You're blending together too many different functions - after all, a warehouse robot that can bring you any pallet in the joint within two minutes isn't going to be able to sweep floors or cover second at the company baseball game, either - but I'll venture an answer that covers all but the extremes of your question.
Between now (with varying levels of success) and five years. None of the core flow of that process requires human-level mentation; it's assumed that a virtual farm of 500 software-development bots are being supervised by some small number of humans who do the creative thinking and go kiss-azz to the board for funding.
No one including me is saying AI will completely, totally replace each worker as they are now defined and employed, but AI will take over large swathes of the more defined and repetitive work, leaving only the fuzzy warm edges for humans. Job classifications and "employee" definitions will change, just as they did with simpler robotic/automation evolution.
Just to bring it all the way to bare concrete: anyone who thinks "AI is going to take over the workforce" means humanoid robots will be sitting in all the current chairs does not understand the nature of the technology or how it will be implemented, any more than assuming "robot welders" would be Marvin with an oxy torch.
Yes, but only if it learns through pattern-recognition, that such an action is expected of it, for some programmed purpose. For AI to actually think about what it's doing - to ponder the concept, the philosophy, the ultimate reason... to actually have the intrinsic desire to do it, for its own personal ends... that is pretty unlikely to be achieved in the 21st century.
What is the "purpose" of a human being?
The "desires" and "personal ends" of AI are whatever its "software" computes it to be. Like humans. Humans do not philosophize or ponder ultimate reason now, with rare exceptions. Even then it is mostly a clever mind devising arguments to support its preconceived notions and feelings.
When AI tech is able to match human processing ability, I think you will surprised at how functionally superior it will be. It will not be scattered, emotional, and torn in a dozen different directions. It will be focused, and perfect in execution. And whatever abilities it acquires can be flawlessly and instantly "taught" to another of its kind.
When do you think that AI would be able to examine several machined parts, realize that the assembly would have less friction and better robustness if it were designed differently... then go through those designs, being mindful of cost, of customer-acceptance, of attitudes of the company executives... coordinate with the subject-matter experts in aerodynamics, in NVH, in acoustics, in drivetrain-development and so forth... run said tests and computations, and finally, weld together a prototype... which it would then demonstrate on the track... and give a convincing presentation to management?
Some of the self appointed experts on this forum would have you believe that AI is able to do all of this and more. Those of us who have worked on the cutting edge of computerization and robotics can get enthused but know better. Computers can do a great job of playing chess were brute force calculations can have an advantage but they are a long way from "thinking" and being able to analyze. Most of the current applications are incredibly crude.
Some of the self appointed experts on this forum would have you believe that AI is able to do all of this and more.
You may be referring to other claims made among the 1000+ posts, but I've never said anything of the kind, and in fact my last handful of posts have been debunking the idea that human-level AI or AGI or anything like it is on the immediate horizon.
However, specialized forms of AI, considerably advanced from what's now in use, will begin replacing a whole new tier of workers within five years. We've already seen the leading edge of it in interactive voice systems that can handle pretty much the whole range of caller concerns, etc. (And which turn the caller over to a real human when its limits are exceeded... just as network admin bots and analyst boxes will.)
I have nothing much more to add about AI itself; it's tangential to my interests and I draw on close association with an expert colleague who's noted for breakthrough thinking in the field.
My interest is entirely this: the number of "good" jobs has been steadily shrinking over the past few decades, from the loss of career jobs in heavy industry, to the bigger loss of career jobs in factory, manufacturing and warehouse segments, and this is significantly due to what can broadly be called "automation" or "robotics" or "efficiency engineering" and now, the leading edge of "AI." The job loss has been masked to some extent by a rise in whole "service" sectors that were either minor or previously did not exist, but we can only sell each other so many backrubs, dining experiences and counseling sessions. We have comforted ourselves with the idea that sweaty workers could just go overseas or to North Dakota or wherever Toyota builds its next US plant, because the rest of us were going to get STEM degrees and run the world (of these sweaty workers) from cool, air-conditioned campuses in groovy parts of the country.
Unfortunately, the next ten years of AI will displace a very large segment of this worker class (taking with it an awful lot of discretionary "service" needs), and pretty well demolish the idea that there is a "good" - lifetime, career, economically viable - job for anyone who wants to exert themselves hard enough to get one. Yes, there will still be STEM jobs, the way there are still plant and warehouse managers who now watch a tireless fleet of 'bots do the job. Instead of 50-100-1000 sweaty workers or Eddie Bauer-wearing desk jockeys.
We need a new economic model, not from political or theoretical or idealistic grounds, but for hard practical survival.
I searched for the term “singularity” in this thread but did not find it. Forgive me if this has been explained already but the thread is over 1000 posts...
Ridiculous? I have no idea. But that number, 2040, catches one’s eye.
It isn't ridiculous if the hardware advances pan out. Currently I'm not real optimistic on that score. The method of making processors is facing some hard physical limitations. Super-intelligence is probably going to need a completely new undiscovered method of computation and as such, is pretty far fetched SciFi at the moment.
I started this thread because I saw that we would face severe social issues due to AI even if the hardware did not become much more advanced than it is now. If we don't fix those issues there is little reason to worry about the Singularity.
Current hardware is at the rodent brain level at best. So what would you expect?
Even then, it's able to exceed human capability at driving a car, playing chess, Jeopardy, Go.
Just go go all diabolus on your argument, AGI will require a genuine breakthrough in processing, not just endless speedup of Babbage's machine. That breakthrough has eluded us just like sustained fusion... but we'll get there.
OTOH, a box running at hundreds of teraflops can use existing processing and programming to do a pretty good job of simulating human efforts... which is all that's needed. The BS-y argument above that a single entity should be able to conceive of worlds, sell the idea and scratch its azz is misdirection. All we need is a box that can replace a network administrator for all but the most extreme situations, a programbot that knows as much as any new CS graduate about software development, etc. And those are on the horizon where they aren't in actual development.
Please register to post and access all features of our very popular forum. It is free and quick. Over $68,000 in prizes has already been given out to active posters on our forum. Additional giveaways are planned.
Detailed information about all U.S. cities, counties, and zip codes on our site: City-data.com.
Please register to participate in our discussions with 2 million other members - it's free and quick! Some forums can only be seen by registered members. After you create your account, you'll be able to customize options and access all our 15,000 new posts/day with fewer ads.
)
