Economic impact of AI on wide range of jobs

[ElmerJFudd]

“Professor Geoffrey Hinton told BBC Newsnight that a benefits reform giving fixed amounts of cash to every citizen would be needed because he was “very worried about AI taking lots of mundane jobs”.

 

“I was consulted by people in Downing Street and I advised them that universal basic income was a good idea,” he said. 
 

He said while he felt AI would increase productivity and wealth, the money would go to the rich “and not the people whose jobs get lost and that’s going to be very bad for society.’” - Professor Hinton is the pioneer of neural networks, which form the theoretical basis of the current explosion in artificial intelligence. Until last year he worked at Google, but left the tech giant so he could talk more freely about the dangers from unregulated AI.

https://www.bbc.com/news/articles/cnd607ekl99o

 

[KenElevenShadows]

I won't join in for the usual AI Hate-A-Thon because everyone is so adept at doing that by now. It's not that I'm not wary of AI. It's just that saying the same concerns over and over is not my area of interest during this discussion.

 

I'll just ask this: what are the chances that any government or society ever puts forth a universal basic income? I think it's just about nil.

[ElmerJFudd]

1 hour ago, KenElevenShadows said:

I won't join in for the usual AI Hate-A-Thon because everyone is so adept at doing that by now. It's not that I'm not wary of AI. It's just that saying the same concerns over and over is not my area of interest during this discussion.

 

I'll just ask this: what are the chances that any government or society ever puts forth a universal basic income? I think it's just about nil.

It’s not so much the notion of a universal income - it’s that the developers of this tech believe that it will be so disruptive that a good percentage of the population will be jobless before governments have a system in place to deal with that.  
 

The other thing the article brings up is warfare.  The most powerful nations in the world are already in a race to develop the best AI for accurate and efficient killing and destruction.  There doesn’t seem to be interest between the most capable nations to discuss rules of AI warfare.  
 

 

[KuruPrionz]

4 hours ago, ElmerJFudd said:

It’s not so much the notion of a universal income - it’s that the developers of this tech believe that it will be so disruptive that a good percentage of the population will be jobless before governments have a system in place to deal with that.  
 

The other thing the article brings up is warfare.  The most powerful nations in the world are already in a race to develop the best AI for accurate and efficient killing and destruction.  There doesn’t seem to be interest between the most capable nations to discuss rules of AI warfare.  
 

 

There will be no "rules" for AI warfare, only destruction. Artificial Intelligence has no moral standards, no qualms regarding taking any sort of actions, etc.

AI Warfare can only be conducted by humans who are sociopathic and will stop at nothing. Those of us with feelings and reservations regarding harm to other humans may be utilized but certainly much less effective for the sort of slaughter required by warfare. 

Best case would be AI vs AI, single digitally coded entities going at it one on one, winner take all. At some point that will happen anyway, remember "Big Time Wresting"?

Same concept only no actual pain in accrued. 

[Anderton]

6 hours ago, KenElevenShadows said:

I won't join in for the usual AI Hate-A-Thon because everyone is so adept at doing that by now.

 

I certainly don't hate AI. I just have zero faith in the humans who will be the custodians of what it can and cannot do. It would take only a handful of really bad people to undo all the good that AI can bring to, for example, medicine.

 

Remember, humans thought it was a good idea to do self-checkout. What could possibly go wrong?   It's not the machines that steal stuff. 

 

To be more honest than I probably should, the percentage of people doing good vs. the percentage of people who aren't probably isn't that much different from the percentage of YouTube comments that are helpful/civil compared to the ones that aren't. When Idiocracy came out, it seemed ludicrous and implausible. Yet here we are...

[KuruPrionz]

Just ran across this, it's a photo but the text is pretty stellar. 

 

[KenElevenShadows]

8 hours ago, ElmerJFudd said:

It’s not so much the notion of a universal income - it’s that the developers of this tech believe that it will be so disruptive that a good percentage of the population will be jobless before governments have a system in place to deal with that.  

 

 

 

I brought up universal basic income because the article brought it up. The professor brought it up. I understand already that the tech will be disruptive. I believe that's been discussed to death, so I tried to bring up something else. 

 

What you are mentioning is the very reason the professor brings up universal basic income.

[KenElevenShadows]

2 hours ago, Anderton said:

 

I certainly don't hate AI. I just have zero faith in the humans who will be the custodians of what it can and cannot do.

 

Exactly. As I keep saying, it's not AI that need truly concern us. It's who is developing or using it.

[KenElevenShadows]

48 minutes ago, KuruPrionz said:

Just ran across this, it's a photo but the text is pretty stellar. 

 

 

That's perhaps reasonably accurate if discussing machine learning scraping and learning images and music. Sort of. At least, a lot of the time.

 

His statement is often wildly inaccurate at describing many other things. 

 

Why doesn't anyone discuss its helpfulness in the following? It can be monitoring various aspects of agricultural elements, detect cancerous cells, alert people when gunshots are heard, plan gardens with drought-resistant plants, aid with health and nutrition plans, design hardware for spaceships and ten million other things that are lighter, better, and stronger than human designs, analyze data sets, sort through archives of pictures, fix bugs in codes, make insulin analogs for diabetics and immune cells that fight cancer or aid in Parkinson's or malaria research and identification, test hypotheses faster than humans, identify diseases in humans and plants, create more effective spam filters for email, cleaning offices, analyze chronic conditions with lab and other medical data to ensure early diagnoses, use  the combination of historical data and medical intelligence for the discovery of new drugs, identify defects and nutrient deficiencies in the soil, improve the in-vehicle experience and provide additional systems like emergency braking, blind-spot monitoring, and driver-assist steering for driving cars, detect changes in financial transaction patterns and other potential red flags that can signify fraud, in astronomy it can establish that galaxy mergers were the primary force underlying starbursts, analyze immense teraquads of data, identify data security threats...

 

Because it makes a crappy meme, that's why.

 

When we express opinions, it has to be "I'm totally against something" or "I'm totally for something." We can't ever be for some aspects, against other aspects, and wary of yet others.

[ElmerJFudd]

4 hours ago, KenElevenShadows said:

 

I brought up universal basic income because the article brought it up. The professor brought it up. I understand already that the tech will be disruptive. I believe that's been discussed to death, so I tried to bring up something else. 

 

What you are mentioning is the very reason the professor brings up universal basic income.

I know the concept of a universal basic income is unpopular with just about anyone that gets up everyday to go to work.  But at the same time, when we reach some percentage of jobs lost what will the solution for providing for oneself and one’s family look like? 

[nursers]

4 hours ago, KenElevenShadows said:

 

That's perhaps reasonably accurate if discussing machine learning scraping and learning images and music. Sort of. At least, a lot of the time.

 

His statement is often wildly inaccurate at describing many other things. 

 

I like me some Chomsky but couldn't agree more - this one is off mark.

[BluMunk]

5 hours ago, KenElevenShadows said:

 

Why doesn't anyone discuss its helpfulness in the following? It can be monitoring various aspects of agricultural elements, detect cancerous cells, alert people when gunshots are heard, plan gardens with drought-resistant plants, aid with health and nutrition plans, design hardware for spaceships and ten million other things that are lighter, better, and stronger than human designs, analyze data sets, sort through archives of pictures, fix bugs in codes, make insulin analogs for diabetics and immune cells that fight cancer or aid in Parkinson's or malaria research and identification, test hypotheses faster than humans, identify diseases in humans and plants, create more effective spam filters for email, cleaning offices, analyze chronic conditions with lab and other medical data to ensure early diagnoses, use  the combination of historical data and medical intelligence for the discovery of new drugs, identify defects and nutrient deficiencies in the soil, improve the in-vehicle experience and provide additional systems like emergency braking, blind-spot monitoring, and driver-assist steering for driving cars, detect changes in financial transaction patterns and other potential red flags that can signify fraud, in astronomy it can establish that galaxy mergers were the primary force underlying starbursts, analyze immense teraquads of data, identify data security threats...

 

Because it makes a crappy meme, that's why.

 

When we express opinions, it has to be "I'm totally against something" or "I'm totally for something." We can't ever be for some aspects, against other aspects, and wary of yet others.

 

At least for me, the reason is that we don't get to have discussions, debate and analysis about whether the possible benefits are worth the possible costs. This has always been true... as a species we plow ever forward towards some short term efficiencies or gains, and only once it's impossible to go backwards (and once advocates/adopters of the new shiny thing have made their money) do we evaluate and mitigate the correlating damage.

 

Maybe everything you mention is a given outcome of this new tech. I'm not convinced that that means the tech is a net good, or that it should be embraced as-is.

[Notes_Norton]

Punch cards put loom operators out of work, and all the Luddites in the world couldn't stop that.

 

Quicken, TurboTx, and others put accountants out of work.

 

Direct Dialing put telephone switchboard operators out of work.

 

Voice to text put stenographers out of work.

 

Self-checkouts put cashiers out of work.

 

Robots put automobile and other assembly line workers out of work.

 

Harvesters put farm laborers out of work.

 

DJs put musicians out of work.

 

Where are the lamplighters, electric meter readers, telephone installers, Linotype operators, typesetters, ice delivery men, telegraph operators, elevator operators, film projectionists, milkmen, camera film developers, soda jerks, ditch diggers, blacksmiths, paperboys, gas pump attendants, pin setters, TV repairmen, highway toll takers, and so many others who have been replaced by technology?

 

To survive, they all had to find something else to do in order to make a living.

 

AI will definitely put people out of work. — It's nothing new. We can all be Luddites about it, but that never worked in the past, probably won't work in the future.

 

Capitalists want to increase their profits, and one way to do that is to lower their labor costs. It's nothing new. Most things seem to be happening at a faster rate than they did in the past, but technology feeds technology.

The question so many think about is this: If they put too many of us out of work, who will have enough money to buy their products?

 

I suppose there will be a balance point. The richest of the rich will profit, some displaced workers will find other jobs, a few in the industries that replaced them, and others will tragically be forced into poverty and hardship.

 

Fortunately for me, as a live musician, my advancing age will eventually put me out of work long before AI takes my job.

 

Notes ♫

[KuruPrionz]

I understand the comments on Noam Chomsky's rant. 

That said, if one is going to elaborate in full detail on all aspects of AI, everybody above has briefly skimmed a variety of other uses.

So, y'all are doing the same exact thing, brief and tiny discussion of possibilities. 

 

One could write a complete essay on ALL aspects of AI, if one wanted to make a complex and boring book about every tedious detail of the technology.

I know I am not going to do that. Feel free to jump in. 😇

[BluMunk]

2 hours ago, Notes_Norton said:

The question so many think about is this: If they put too many of us out of work, who will have enough money to buy their products?

 

I suppose there will be a balance point. The richest of the rich will profit, some displaced workers will find other jobs, a few in the industries that replaced them, and others will tragically be forced into poverty and hardship.

 

Fortunately for me, as a live musician, my advancing age will eventually put me out of work long before AI takes my job.

 

 

Quite true... in the medium- or long-term the economics will take care of itself. There's just potential for serious misery in the short term as people's lives are disrupted.

Also, the difference between AI and the examples you listed is that AI will quickly impact a vast number of industries incredibly quickly. We can absorb/tolerate a small number of people losing their jobs/communities loosing employers for the sake of innovation, but I am doubtful that we can smoothly transition the vast numbers whose jobs are likely going to disappear in the upcoming decade.

 

[Anderton]

6 hours ago, ElmerJFudd said:

I know the concept of a universal basic income is unpopular with just about anyone that gets up everyday to go to work.  But at the same time, when we reach some percentage of jobs lost what will the solution for providing for oneself and one’s family look like? 

 

Buckminster Fuller estimated that if every single person spent two full weeks of the year maintaining or creating what was needed for the earth's population (food, transportation, energy, medical care, etc.), then there would be enough to provide everyone with food, clothing, shelter, and medical care for a year. Then, you wouldn't have to be concerned about surviving. If you just wanted to play guitar all day for the other 50 weeks, you could. But if you wanted to create an electric car company and become a billionaire, you could do that too. Of course, this would require a immensely coordinated global effort, which would never happen given the fractious nature of earth people. But he was pretty sure that if humanity had the will, it was doable. 

 

It would help the concept considerably if some of his ideas for energy were implemented. He pictured a global grid where the lines went down from sustainable energy creation at the poles - one running down the Americas, one through Europe and Africa, one West Asia and the Middle East, one East Asia and Oceania, and with load balancing because half of the earth would be in sunshine while the other half was in shadow...that kind of thing. 

[K K]

15 hours ago, KuruPrionz said:

There are will be no "rules" for human AI warfare, only destruction.

 

Fixed, as this is what is already happening since decades and right now as well.

[KenElevenShadows]

On 5/20/2024 at 4:30 AM, nursers said:

I like me some Chomsky but couldn't agree more - this one is off mark.

 

Same. I like the guy, but he's super off the mark here. But it makes a good meme, and that's what matters now.

[Stokely]

Quote

the money would go to the rich

 

There's the part I'd focus on, not just about AI.  But of course you can't bring up wealth inequality--how it's trending, what it might mean--without rapidly getting political.

At some point UBI is going to be needed if it isn't already.

 

[Notes_Norton]

It's coming, a thing called A I

The people are saying, “Oh my!”

Our jobs will be gone, we're losing our home

And now we are all gonna die.

 

(This limerick was NOT generated by AI — AI probably would have written a better one)

 

 

Notes ♫

 

 

 

[KenElevenShadows]

I asked Gemini AI to compose a quick poem, Notes. Here's its first attempt.

 

The robots are coming, their circuits all hot,

To steal all our jobs, from barista to yacht.

No more flipping burgers, no more coding a line,

Just AI overlords serving lukewarm wine.

 

But fear not, dear reader, their reign will be brief,

For without human chaos, their systems will grief.

They'll crash in confusion, their plans all askew,

When nobody argues which ketchup is "true".

[David Emm]

On 5/20/2024 at 12:46 PM, Anderton said:

Buckminster Fuller estimated

 

Fuller is one of my weird heroes, because he once said, IIRC, "I have made more mistakes than any others of whom I know, yet those mistakes are what enabled me to eventually devise more useful instrumentalities for addressing complex issues." Excellent. Its like old Yes lyrics, except that it makes sense. 

[David Emm]

1 hour ago, KenElevenShadows said:

I asked Gemini AI to compose a quick poem, Notes. Here's its first attempt.

 

The robots are coming, their circuits all hot,

To steal all our jobs, from barista to yacht.

No more flipping burgers, no more coding a line,

Just AI overlords serving lukewarm wine.

 

But fear not, dear reader, their reign will be brief,

For without human chaos, their systems will grief.

They'll crash in confusion, their plans all askew,

When nobody argues which ketchup is "true".

 

I respond with an ancient bit of lore from the comic strip "B.C.":

 

Most of my poems are written in haste

and are, therefore, resultantly, lacking in taste

Yet people who read them and think they are fine

Must surely have taste just as rotten as mine.

[Anderton]

19 hours ago, David Emm said:

"I have made more mistakes than any others of whom I know, yet those mistakes are what enabled me to eventually devise more useful instrumentalities for addressing complex issues."

 

Check out the last paragraph on page 42 of the latest issue of Mixonline...

[Philbo King]

I fear human stupidity far more than AI.  I have no use for AI, but I think AI will be a human stupidity multiplier or more likely an exponentiator.

[ElmerJFudd]

13 hours ago, Philbo King said:

I fear human stupidity far more than AI.  I have no use for AI, but I think AI will be a human stupidity multiplier or more likely an exponentiator.

Of course.  We made the Ai and are slow to deal with the ramifications because the only ones who see where this is going are those in the industry, those investing in the industry and those that care to pay it a thought.  

[Anderton]

Then again, humanity's best defense against AI is artificial stupidity. We can make really stupid mistakes but in the process, discover something cool by accident. So far, machines can't do that.

[KenElevenShadows]

On 5/24/2024 at 9:29 AM, Anderton said:

Then again, humanity's best defense against AI is artificial stupidity. We can make really stupid mistakes but in the process, discover something cool by accident. So far, machines can't do that.

 

Self-driving automobiles operated by AI were having difficulty identifying black pedestrians crossing the street at one point (hopefully not now). Does that qualify as "artificial stupidity"?

 

Well, regardless, it's not discovering something cool by accident, so there is that.

[Notes_Norton]

8 hours ago, KenElevenShadows said:

 

Self-driving automobiles operated by AI were having difficulty identifying black pedestrians crossing the street at one point (hopefully not now). Does that qualify as "artificial stupidity"?

 

Well, regardless, it's not discovering something cool by accident, so there is that.

I think eventually self-driving automobiles will be safer, but until version 4.0 or 5.0 comes around and gets all the bugs out, there will be deaths. 

 

I see a lot of naturally stupid drivers every day.

 

[KenElevenShadows]

5 hours ago, Notes_Norton said:

I think eventually self-driving automobiles will be safer, but until version 4.0 or 5.0 comes around and gets all the bugs out, there will be deaths. 

 

I see a lot of naturally stupid drivers every day.

 

 

Agreed. There will come a day in which the self-driving automobiles are super great. They're allegedly statistically safer than humans as it is (which according to my tech friends is more of a condemnation of how poorly humans drive than it is how well self-driving automobiles fare), but when they don't know what to do, they just stop.

 

But they are still in their infancy, and can only get better from here.

 

I think it they would be even safer if all cars were self-driving cars. I have no actual hard evidence to back that up, of course, but it seems logical.