U.S. CitiesCity-Data Forum Index
Go Back   City-Data Forum > General Forums > Economics
 [Register]
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.
View detailed profile (Advanced) or search
site with Google Custom Search

Search Forums  (Advanced)
 
 
Old 08-07-2018, 03:27 PM
 
Location: Arcadia, CA
101 posts, read 32,196 times
Reputation: 148

Advertisements

Quote:
Originally Posted by raggedjim View Post
Jobster, Boston Dynamics has done some interesting things, and I do believe changes are coming. I'm just not convinced it will be in 5 years, I think it's more of a 15-20 year time-frame.

Good luck to all of us, we will need it, Rg
I also think 5 years are too quick. Given human attraction to predictability and familiarity, I don't think business would begin employing AI en mass as soon as it is commercially available.
Quick reply to this message

 
Old 08-07-2018, 03:57 PM
 
1,461 posts, read 330,127 times
Reputation: 1663
Quote:
Originally Posted by Quietude View Post
I believe we were arguing somewhere between intractably and pointlessly about this a few months ago.

https://www.nytimes.com/2018/08/05/t...kplace-ai.html

Anyone who still believes the lower third of traditional white-collar desk jobs are secure in the hands of human employees, and not on the verge of being replaced by simple AI - and to forestall the stupider arguments, I emphasize simple - is simply denying that the waves that wiped out whole job categories in heavy industry, then warehousing and manufacturing are breaking yet higher.

I predict, with a greatly conservative take, that the impacts of "office drone" AI will be painfully sharp within five years. The employment in junior accountants, actuaries, order processing, similar repetitive/low judgment positions, and increasing areas of customer support, will fall like dominoes.

But hey, they can all go teach yoga or give pedicures or arrange flowers or something, right?
I am a network engineer and I can attest to many duties that have been chipped away, which once commanded much more staffing at telephone companies than prior years:

- Auto attendant: call an 800# and get a touch tone menu to go directly to sales, directly to service, or check your balance without even speaking to a human.
- Multiple call paths: high volume centers can handle several simultaneous calls going through the same menu, whereas a human can take one call at a time, the rest remaining on hold.
- When business customers call in a down circuit, there has already been a ticket created, and the circuit has already been tested. For those not choosing to use online portal, by the time they call in, we've already dispatched a technician.
- Certain conflicts in automation (especially when it's crossing platforms) do need to get escalated to application engineers, but these are few and far between.

In other areas, like credit cards, so long as your score and income meet automatic approval requirements, the process is almost entirely automated. You are even given your APR and credit limit in the instant result.
Quick reply to this message
 
Old 08-07-2018, 06:24 PM
 
Location: Aurora Denveralis
2,978 posts, read 1,012,279 times
Reputation: 3796
Yes, yes. Either "we've been warned for years and it's never happened" or "it's been happening forever and we've adapted." Got it.

Both groups are completely failing to understand the ground shift that's about to happen. The vague hand-flapping argument that there will somehow be as many kiosk-servicing jobs as there were counter service jobs is just the beginning of the misunderstanding. That "mechanical" jobs have always been subject to mechanization/automation is another. That formerly inviolate, human-intelligence-only jobs are now on the chopping block is the key change. This isn't eliminating the typing pool because the boss has Word (and probably still has juniors to do his typing, anyway); this is the elimination of whole job categories that have been regarded as un-replaceable, and the dominoes will fall faster and further than almost anyone can presently grasp.

See youse in five years or so. If you can still afford internet access.
Quick reply to this message
 
Old 08-07-2018, 06:26 PM
 
24,885 posts, read 11,599,261 times
Reputation: 11617
Quote:
Originally Posted by lieqiang View Post
This. The "technology will kill all the jobs" hand wringers don't seem to grasp that human ingenuity has been replacing people with tools and machines for centuries.

In five years, the timeline of OP's great dive of office worker employment, the number of employees in the field will be driven far more by the state of the economy itself and not increases in productivity.

The people claiming that it will create new jobs don't seem to grasp why this is different. Automation in the past allowed us to move into things where we excelled at that automation could not do.

With AI and advanced robotics we are going to move into situations where quite literally we can replace us entirely and we cannot compete in any meaningful way. We wont be more creative, or capable then what we can make.

Its like horses and wheels. wheels allowed horses to add new abilities such as carriages etc. The steam engine took over long distance travel.....but then the automobile showed up and horses are now primarily pets. This time whats coming is fundamentally different then in the past.
Quick reply to this message
 
Old 08-07-2018, 06:28 PM
 
1,461 posts, read 330,127 times
Reputation: 1663
Quote:
Originally Posted by Quietude View Post
Yes, yes. Either "we've been warned for years and it's never happened" or "it's been happening forever and we've adapted." Got it.

Both groups are completely failing to understand the ground shift that's about to happen. The vague hand-flapping argument that there will somehow be as many kiosk-servicing jobs as there were counter service jobs is just the beginning of the misunderstanding. That "mechanical" jobs have always been subject to mechanization/automation is another. That formerly inviolate, human-intelligence-only jobs are now on the chopping block is the key change. This isn't eliminating the typing pool because the boss has Word (and probably still has juniors to do his typing, anyway); this is the elimination of whole job categories that have been regarded as un-replaceable, and the dominoes will fall faster and further than almost anyone can presently grasp.

See youse in five years or so. If you can still afford internet access.
Both remarks are frighteningly dismissive towards an issue that will ultimately result in loss of income to naysayers who refuse to acknowledge a threat.
Quick reply to this message
 
Old 08-07-2018, 08:09 PM
 
394 posts, read 244,753 times
Reputation: 1461
The day an AI can do what I do, I will deserve to lose my job. It'll have to be one sharp, quickly adaptive, trouble-shooting, problem-solving, fire-stomping s.o.b. that never knows what it will be doing from day to day.

Good luck, Rg
Quick reply to this message
 
Old 08-07-2018, 09:09 PM
 
6,815 posts, read 4,408,035 times
Reputation: 11918
Yes, this time it's different. Got it.

I wholeheartedly accept the challenge of revisiting the topic in 5 years. And 5 years thereafter. And 5 years after that. And so forth.

Meanwhile, I'm going back to refining lecture notes for a graduate class in experimental methods in aeronautics, which I may be teaching soon, at the same university in which I took the same class 25 years ago. Back then, the notes were based mostly on material from the 1960s. I do not expect to be making extensive changes. I also expect that essentially the same notes will be used in 2060, albeit perhaps in a different delivery-format... and, as some here seem to think, by an AI lecturer.
Quick reply to this message
 
Old 08-08-2018, 10:39 AM
Status: "delete" (set 20 days ago)
 
3,189 posts, read 1,273,221 times
Reputation: 2351
Quote:
Originally Posted by ohio_peasant View Post
Yes, this time it's different. Got it.

I wholeheartedly accept the challenge of revisiting the topic in 5 years. And 5 years thereafter. And 5 years after that. And so forth.

Meanwhile, I'm going back to refining lecture notes for a graduate class in experimental methods in aeronautics, which I may be teaching soon, at the same university in which I took the same class 25 years ago. Back then, the notes were based mostly on material from the 1960s. I do not expect to be making extensive changes. I also expect that essentially the same notes will be used in 2060, albeit perhaps in a different delivery-format... and, as some here seem to think, by an AI lecturer.
It is unlikely that there will be any fundamental changes to some sciences; however, given the vast amount of data and the ability to rapidly process this information into something meaningful may mean that some anomalies are discovered that do lead to some significant changes or future discoveries.

AI can't alter physics. It still has to operate in our world of limitations. AI is ultimately reliant on data. The more data available, the more correlations that can be drawn from that data to formulate conclusions and further increase the knowledge base.

What is your expectation for AI? I just showed in a previous post on this thread where deep learning processes will be utilized to create an autonomous drone, utilized for identifying and potentially hunting humans. The project is currently scheduled to be online as early as the end of the year, so I'm not sure why you used aerospace as an example, as its clearly an industry that is practically automated as it already is.
Quick reply to this message
 
Old 08-08-2018, 05:14 PM
 
Location: Vallejo
13,437 posts, read 15,036,253 times
Reputation: 11924
Quote:
Originally Posted by Quietude View Post
I believe we were arguing somewhere between intractably and pointlessly about this a few months ago.

https://www.nytimes.com/2018/08/05/t...kplace-ai.html

Anyone who still believes the lower third of traditional white-collar desk jobs are secure in the hands of human employees, and not on the verge of being replaced by simple AI - and to forestall the stupider arguments, I emphasize simple - is simply denying that the waves that wiped out whole job categories in heavy industry, then warehousing and manufacturing are breaking yet higher.

I predict, with a greatly conservative take, that the impacts of "office drone" AI will be painfully sharp within five years. The employment in junior accountants, actuaries, order processing, similar repetitive/low judgment positions, and increasing areas of customer support, will fall like dominoes.

But hey, they can all go teach yoga or give pedicures or arrange flowers or something, right?
Nothing new. In my field there used to be a 2:1 ratio of clerical workers. It's now more like 1:5. 90% of the clerical typists and secretaries have just been replaced by photocopy machines and computers. Most of the AI happened in the '90s.
Quick reply to this message
 
Old 08-09-2018, 10:59 AM
 
6,815 posts, read 4,408,035 times
Reputation: 11918
Quote:
Originally Posted by Jobster View Post
...What is your expectation for AI? I just showed in a previous post on this thread where deep learning processes will be utilized to create an autonomous drone, utilized for identifying and potentially hunting humans. The project is currently scheduled to be online as early as the end of the year, so I'm not sure why you used aerospace as an example, as its clearly an industry that is practically automated as it already is.
You have asked a serious question, so I’ll attempt a serious reply.

First, it seems to me that the physical sciences and their application to research, have attained maturity by the mid 20th century. Innovations since then, however substantial, have been mostly in miniaturization, growth in speed/power/efficiency and so forth, rather than qualitative innovations in the thing itself. Recall that the digital computer, the laser, micro-electronics, supersonic flight, nuclear power and so forth, were already mature by the 1960s. 50 years later, we still regard some or all of those things as “high tech”. But step 50 years back from the 1960s, which would be the WW1 era, and those ideas would have been utter fantasy, rather than some abstract concept whose practical advent would could soon anticipate. By this I mean that in a certain sense the pace of innovation has slowed down, rather than accelerated.

Second, we humans, especially in the first world, have a remarkable talent for creating busy-work, keeping employed persons without much acumen, talent or drive, but who somehow go about perfunctory duties in vast administrative apparatus. A key example is the healthcare industry, where so few of the actual participants develop drugs or treat patients or perform lab-tests. Most are involved in paperwork, insurance, administration, compliance with regulations, litigation, facilities maintenance or whatever else. In my workplace, the secretaries are almost entirely gone. Engineers type their own reports on their laptops. The graphic arts department is also almost entirely gone. Engineers do their own presentations in Powerpoint, and make their own display-posters in Photoshop. But persons formerly in secretarial duty or in graphic arts, still hang around. Now they are security specialists, or event coordinators of some type, or minor financial functionaries.

Thus, as robots and computers start taking over more and more essential duties, obviating humans, the humans would carry on with non-essential duties. This is possible because the more efficient the robots become, the more the human employment can be carried as overhead.

I am reminded of the prophetic essay by John Maynard Keynes, written in 1930, about a life-of-leisure awaiting humans a century hence. See for example http://www.econ.yale.edu/smith/econ116a/keynes1.pdf. Perhaps that essay deserves its own thread? Was Keynes fundamentally wrong? If so, how?

As to your example of AI in drones etc., I can say as an “insider” in the profession, that this stuff has already been around for a decade or longer. Its widespread implementation was going to be 1-2 years away, for the past 10 years. Why the delay? It’s because the availability of a technology does not imply acceptance. The lag between the two is, if anything, increasing. Society is becoming more risk-averse and more conservative. The FAA has spent over a decade, diddling with rules about usage of drones (back when they were still called UAVs) in public air space. To my knowledge, this “process” is still unfolding. And this is about plastic helicopter with dumb VGA cameras aboard – no AI. Amazon famously proposed home-package delivery by drone. What happened? Has anyone actually received a package by such means? I would not be surprised if in 2030, 2040, even 2050, Amazon’s packages are still delivered by truck, driven by a human.
Quick reply to this message
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 update this thread with any new information or opinions. This open thread is still read by thousands of people, so we encourage all additional points of view.

Quick Reply
Message:
Over $104,000 in prizes was already given out to active posters on our forum and additional giveaways are planned!

Go Back   City-Data Forum > General Forums > Economics
Similar Threads
Follow City-Data.com founder on our Forum or

All times are GMT -6.

2005-2018, Advameg, Inc.

City-Data.com - Archive 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35 - Top