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)
Reply Start New Thread
 
Old 04-03-2019, 10:43 AM
 
1,057 posts, read 226,373 times
Reputation: 1809

Advertisements

Ginni Rometty, CEO of IBM, was on CNBC this morning warning about AI affecting everyone's job in the future.

Of course, the idea that machines are coming to take our jobs has been around for hundreds of years, and it has always been wrong. "This time will be different," say the AI scarecrows.

This brings to mind the classic story of the invention of the automatic knitting machine back in the 16th Century. At a time when technology in the textile industry meant spinning wheels and hand looms, clergyman William Lee invented a machine that could knit stockings.

In 1589, Queen Elizabeth I of England was alarmed when Lee applied for a Royal Patent for his knitting machine, and said:

Quote:
"Consider thou what the invention [of the automatic knitting machine] would do to my poor subjects," she pointed out. "It would assuredly bring them to ruin by depriving them of employment and thus make them beggars.”
The above is from "History of the Framework Knitters" written by Gravenor Henson back in 1831 in England. Hensen's thesis was that hosiery, lace and all other industries should be regulated by the government so as to maintain a decent living standard for the workers and fair conditions of trade, that British industries must be protected from direct foreign competition and, more particularly, from industrial espionage, migration of skilled workmen to other countries, and export of machinery.

It didn't quite turn out that way. Productivity went up, GDP went up, the standard of living went up, and there was no widespread unemployment or starvation as a result of automatic weaving and knitting machines. Just the opposite, of course: more people were raised out of abject poverty as a result of the disruption as far more jobs were created than destroyed.

Everyone agrees this is the way it worked throughout history regarding technical innovation and disruptive technologies: society is better off.

Yet somehow, AI scares people (most of whom have no understanding of technology and indeed lack the capacity to spell AI). People just don't get what AI is and does. Journalists write about the dangers of AI but they have zero domain knowledge and have even less subject matter expertise.

At its basic level, AI is good for determining "this is a cat" and "that is not a cat" when examining two photographs, such as:

https://en.wiktionary.org/wiki/cat#/...File:Cat03.jpg

and

https://assetstorev1-prd-cdn.unity3d...31a89989c1.jpg





And somehow, this is supposed to lead to massive unemployment?

Last edited by RationalExpectations; 04-03-2019 at 11:00 AM..
Reply With Quote Quick reply to this message

 
Old 04-03-2019, 01:54 PM
 
4,191 posts, read 2,395,182 times
Reputation: 9598
AI robots now flip McDonald's burgers and serves them up at some locations. Does it make for a cheaper burger? Not so much.
No, not for me id have as soon a person assembled my burger and served it to me.

Machines of all sorts do a lot already. I think it takes only 3 people to operate a soda bottling works. One for each shift.

1984's " go down to the fence and communicate with your neighbor for an hour" said the computer. If AI takes over there will be less human interaction and humans are social animals.

That is the fear.

I remember well the fear that people at the store i worked at had over computer registers that read the bar codes to tally up the total due, handle cc transactions etc.They were scared to death.
Having heen raised in an IBM family, i was exposed to computers as a teen. So when the PC came out, i soon after bought one of my own ( an IBM clone)...$2700 worth of clone PLUS peripherals costs.
So i took my computer to work one day and sat everyone down in front of it, let them play games and let them write letters. The fear was gone soon, at least at my store .

I dont cotton to a robot to do my vacuuming more than just the novelty of it. Same with the burger. And i highly doubt i will want to watch the news with a robot reading the reports of news. I will resit invasive "smart house" tech for as long as i can. I DON'T need an Alexa.

Im not sure the fancy "extras" on cars or self drivers are any great thing either. Ive seen knee jerk reactions and over correcting when a lane departure alert goes off. Not so fun. And self drivers will crash just as well as a human driver. They are already arguing over whom to blame in a self driver crash.

The more it costs to build, the costlier it is to repair or replace. They say it costs a small fortune to replace "smart bumpers" on car, a fender bender can cost as much to repair as a new car or more.

But you might be right, it will just be DIFFERENT jobs will be created but i think we are a long way away from that.

AI can only do so much and humans will only tolerate so much. At least until it becomes main stream.

And AI can do a LOT more than " this is a cat, this is not a cat".

Its when AI starts "thinking" about what animal it is that scares people.

People will have to adjust, thats all. Some slower than others.

Reply With Quote Quick reply to this message
 
Old 04-03-2019, 04:39 PM
 
Location: Ohio
19,070 posts, read 13,871,204 times
Reputation: 15291
Quote:
Originally Posted by RationalExpectations View Post
And somehow, this is supposed to lead to massive unemployment?
It's greatly over-exaggerated by the not-too-bright.

It's claimed that 69% of lawyer, paralegal and legal administrative assistants will be eliminated, which is total crap, because the actual figure is closer to 3%.

I'm all on board for AI in the legal field, because it will give a greater number of people access to legal services. Right now, maybe 20% of households have access and the other 80% don't.

Estate planning? 80% of households can't do it, because they don't have the money. There's lots of legal services people could use, but can't, because they don't have the money.

You really should consult with a real estate attorney, before buying a house or any property, but most people can't afford it.

Who wants to buy a house only find out 7 years later that half their front yard is gone, because the county widened the road to two lanes in both directions with a turn lane in the middle?

Your real estate agent wouldn't know that, and probably wouldn't tell you even if they knew in advance, because they want to make a sale, but a real estate attorney will tell you.

How would they know?

You have to understand why things work the way they do.

The federal EPA, and most State EPAs require environmental impact studies to be done for a great many things, including road/highway construction, commercial or industrial development, airport expansion, zoning law changes and much more.

Your county will publish bid specifications for the purpose of contracting an environmental or engineering firm to conduct the environmental impact study.

And, even before that, your county will publish bid specifications to hire an engineering firm to do a feasibility study to see if it's even possible to widen the road, how long it would take, and the cost projections.

That's all 5 to 10 years before the actual work starts.

Neither your real estate agent nor your finance company would know that, but a real estate attorney would, because they're connected to all the right sources.

Also, note that county tax assessors have a habit of devaluing property they know they will be purchasing in the future through Eminent Domain.

So, you buy your home, then two years later the county tax assessor comes round and devalues your property $30,000 putting you underwater on your mortgage and then three years after that, the county comes round wanting to acquire part of your property under Eminent Domain to widen the road and they're not going to pay you full value, unless you fight them in court, so you get screwed again.

Sucks to be you.

It's a good thing if more people have access to affordable legal services.

Anyway, there are severe limitations on AI in the legal field.

Take court appearances. A computer is not going to show up in court for you.

The same is true for negotiations. In tort law, your attorney will always attempt to negotiate a settlement with an insurance company or other defendant, because legal action is always the last resort. Negotiations are also a large part of business and contract law, and also bankruptcy law, like creditor meetings.

AI is simply incapable of negotiating anything. Maybe 150 to 200 years from now, but not now, and not for anytime in the foreseeable future.

Fact investigation is something else AI is simply incapable of doing. That's also true for advising, communicating and developing legal strategy.

Legal research, legal writing and document drafting are all areas where AI fails miserably.

If you doubt, look at IBM's Watson and the debate thing. It requires human interaction. If AI requires human interaction, then it's not replacing jobs.

Where AI does have at least some impact is document production, also known as "coding."

In civil litigation, you can have a few dozen pages of documents up to Millions of pages of documents. If you remember the savings and loan scandal, there were like 172 Million pages of documents. I worked on a patent lawsuit that had over 3.7 Million pages of documents.

Can you imagine reading through 100,000 pages of documents? Okay, now what? They're of little value to you, unless you can categorize and classify them, and then sort them by relevance. That's the purpose of document coding. You have a database with digital images that uses OCR plus entered data that allows you to quickly find those documents that are relevant and ignore those that are totally irrelevant.

The problem with AI is it cannot pick up on subtle references.

Take a class-action lawsuit against an employer for race or gender discrimination.

A memo from HR says do not promote this person, because she's a woman, but the memo doesn't phrase it as "do not promote this person because she's a woman."

What the memo actually says is "We need team-oriented players focused on the project who do not have other burdens."

AI would never figure it out in a Million years. AI just can't think like a human. It might come close in approximation, but it will never be 100%.

AI is always going to require human interaction, and it's impact on jobs is greatly over-exaggerated.
Reply With Quote Quick reply to this message
 
Old 04-03-2019, 09:15 PM
 
Location: Cebu, Philippines
3,750 posts, read 1,366,677 times
Reputation: 6998
Banks have been running on nearly all AI for decades, but still need sky-scrapers full of employees, none of whom do long division with quill pens in green visors anymore.

We have regulatory agencies to make sure the entire work force is kept busy. Nobody buys newspapers or magazines anymore, and business records are kept digitally, but somehow paper consumption keeps rising. Every Rx is handed to me with 12 sheets of government-mandated sheets of paper stapled to it, which AI is powerless to reduce. All created by five pharmacy assistants, not by Mr. Walgreen, who wrote it down in a ledger.
Reply With Quote Quick reply to this message
 
Old Yesterday, 07:32 AM
 
4,427 posts, read 2,424,216 times
Reputation: 7853
AI is good for certain things but it has many limitations. In fact AI can make things worse and that is exactly what it’s doing currently for my employer.
Reply With Quote Quick reply to this message
 
Old Yesterday, 08:22 AM
 
4,641 posts, read 10,146,425 times
Reputation: 2166
For every robot or AI machine there is another person needed to repair it, a company to make parts & then delivery of said parts.... it goes hand in hand.
Reply With Quote Quick reply to this message
 
Old Yesterday, 08:28 AM
 
13,033 posts, read 6,841,035 times
Reputation: 24045
There's some confusion in this thread between plain-vanilla automation & computerization vs artificial intelligence. If I use my smartphone to photograph and deposit a paper check into my checking account, that's not AI. If I use a kiosk or a smartphone application to order fast food, that's not AI. A machine that cooks burgers doesn't have any AI in it.


Autonomous automobiles, when they finally get here, require AI. They use visual (and/or RADAR) perception to learn and make decisions. The CNC machine in the factory that replaced dozens of machinists doesn't have artificial intelligence.
Reply With Quote Quick reply to this message
 
Old Yesterday, 12:27 PM
 
Location: Aurora Denveralis
5,870 posts, read 2,072,590 times
Reputation: 9072
Quote:
Originally Posted by GeoffD View Post
There's some confusion in this thread...
Yep.
Reply With Quote Quick reply to this message
 
Old Yesterday, 02:46 PM
 
Location: Ohio
19,070 posts, read 13,871,204 times
Reputation: 15291
Quote:
Originally Posted by rocafeller05 View Post
For every robot or AI machine there is another person needed to repair it, a company to make parts & then delivery of said parts.... it goes hand in hand.
That's exactly right.

Every robot-welder that replaces a welder creates 1.3 jobs.

So, for every three robot-welders, you create four jobs.

No one is being left out of a job.

Years from now, we can laugh at the ninnies who mussed and fussed over technology.
Reply With Quote Quick reply to this message
 
Old Yesterday, 03:31 PM
 
Location: Aurora Denveralis
5,870 posts, read 2,072,590 times
Reputation: 9072
Quote:
Originally Posted by Mircea View Post
That's exactly right.

Every robot-welder that replaces a welder creates 1.3 jobs.
This is the kind of smug reductive reasoning that makes me long for our robot masters to take control.

So if twenty welding stations on an assembly line are replaced by automated welding systems (which will worth 3 shifts a day and more) you're saying that creates... 26 jobs? Now, is that replacing 20 shift workers... or 40... or 60... not counting the helpers needed at many manned stations for things like positioning and parts handling? (Parts picking, positioning and clamping are integral to many "welding" systems for such fixed use.)

Or are you saying that for those 20 stations, there will be 26 builders (on other assembly lines) plus programmers and maintenance workers... who will only build those 20 systems instead of hundreds to thousands of them? What's the job ratio in that case?

Or are you just smugly pulling outdated, ill-thought figures from studies that assume all of this will happen in such tiny instances it will have no effect on the greater field and will require great heaps of other employment to keep those trial implementations working?
Reply With Quote 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.


Reply

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.

2005-2019, Advameg, Inc. | Please obey Forum Rules | Terms of Use and Privacy Policy

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