953 replies to this topic

[Elus]

I'm starting this thread because I think we're on the verge of a revolution, the likes of which we've never seen before.

In this thread, I will share articles, links, and updates which discuss and provide evidence for the notion robots are about to cause a massive job crisis of surreal proportions.

http://www.youtube.com/watch?v=cXQrbxD9_Ng&feature=youtu.be

The notion that robotics/AI/software will take human jobs may seem radical at first glance. However, consider the following facts:

• IBM Watson crushed world Jeopardy champions in 2011, demonstrating its capacity for understanding natural language. [1]
• IBM Watson is on the verge of passing a US board medical exam, and IBM intends to use Watson for diagnosis of disease, a task normally dedicated to human doctors. [2]
• Google has created a self-driving car that has had zero accidents for 500,000 miles. [3]
• Millions of people regularly utilize SIRI, an artificial intelligence iPhone assistant, for a variety of tasks ranging from scheduling appointments, information queries, messaging, settings alarms and reminders, getting directions, checking stocks, and many other things.
• Baxter, Rethink Robotics new artificially intelligent robot, is capable of doing tasks normally done by Chinese workers for less than it would cost to employ an actual Chinese person. [4]
• Petman, Boston Dynamics' humanoid robot, is capable of walking upright like a human over rocky terrain. They have also created surprisingly agile quadruped robots. [5][6]

http://www.youtube.com/watch?v=tFrjrgBV8K0

• Google and other laboratories have demonstrated remarkable advances in machine learning, speech recognition, and computer vision [6][7]. Google has plans for a trillion node neural network well under way, which they expect will exceed current their current machine learning capabilities.[8]
• In the US, median income has become decoupled from productivity. While productivity has continued to surge upward, median household income has stagnated. This decoupling began in the early 1990s [9]

Here is a selection of articles on the topic of robots and AI taking human jobs, with more to come:

• Robots are going to take your job (April, 2013)
• Are robots hurting job growth? (September 8, 2013)
• Welcome, Robot Overlords. Please Don’t Fire Us? (June 14, 2013)
• Better Than Human: Why Robots Will — And Must — Take Our Jobs (December, 2012)

Edited by Elus, 21 October 2013 - 03:58 AM.

[niner]

I think that a world in which machines are better than humans at literally everything (yes, even that...) is inevitable. Along with rethinking the concept of the inevitability of death, we are going to have to craft a society where existence does not hinge on having a job, since most jobs will no longer exist. Does that mean socialism? Well, I'm not proposing that we give everyone a solid gold toilet, but a society where 0.01% of the population controls all the wealth, and most of the rest are desperately poor will not last long. Maybe the 0.01% will just decide to kill all the other people, or let them die. The future will be interesting...

[lemonhead]

I've tried to think of solutions, but I'm not up to the challenge.

An arts and crafts based economy, with social status determined by how much hand-made stuff you have?

A gambling and gaming economy? That might appeal to more people.

A minimum guaranteed income would have to be provided in either case.

[okok]

We're still greedy animals. A minority controlled the majority of wealth throughout history. Globalization just made it worse. It'll probably still take quite some time and a lot of wealth generation via automatisation for the 'trickling down' to take effect and making human labor exploitation comparably uninteresting. I don't think there will be another french revolution - most of the population is above the basic needs threshold. Despite the general sense of injustice, it's business as usual after the crisis.

[niner]

I don't think there will be another french revolution - most of the population is above the basic needs threshold. Despite the general sense of injustice, it's business as usual after the crisis.

I don't know. If 60-90% of the population don't have jobs, I don't see how it can be business as usual. If we can't provide that large body of people with reasonable options, I wouldn't be surprised to see them rise up- as in another French revolution. Alternatively, maybe the 0.01% will implement the Skynet solution, and send killer bots out to eliminate the bothersome unnecessary humans. I think that this is going to be one of the toughest problems that humans will face in the 21st century. It's not really hard- there are a number of possible solutions, but our deep-seated ideologies will stand in the way. That's the hard part.

[xEva]

please! The same was said about the first machines back in XVIII and the same was said about computers in XX. And while it is true that, say, in the accounting business, accounting assistants became unnecessary, their jobs went into development of the accounting software, its maintenance and support. Absolutely the same applies to all robots and all automation. Real people, with real jobs and support for those jobs, also administered by people, is what behind it all. It has always been this way and it will always be this way. Those who say otherwise are... not that bright

[lemonhead]

It has always been this way and it will always be this way.

We'll see. Keep an eye on the labor force participation rate. I'll check back with you in 20 years.
In the mean time, try reading The Lights in the Tunnel.

It's not really hard- there are a number of possible solutions, but our deep-seated ideologies will stand in the way. That's the hard part.

Isn't that always the hard part?

There's also the stratification issue; there has to be some opportunity for social mobility. Then there's the feeling of value or self-worth that (sometimes) comes from work,
"I don't like work — no man does — but I like what is in work — the chance to find yourself. Your own reality — for yourself, not for others — what no other man can ever know." - Joseph Conrad, Heart of Darkness

Vonnegut provides an examination of these issues in Player Piano.

[xEva]

oh come on guys. the title of this topic should be
Employment crisis: Robots, AI, & automation will take most menial human jobs


just as it has always been. We are technological species and robots and automation only reflect the growth of technology. Or do you think that robots will program robots in foreseeable future? Perhaps you go as far as wars between men and machines? How very unoriginal and passe

Edited by xEva, 23 October 2013 - 05:23 PM.

[PWAIN]

xEva, you are basing your comments on the past. Mechanical machines replacing human labour. The lack of sentience in these machines all but assured continued employment for humans. The problem is when we get machines with advanced AI that makes them effectively indistinguishable from humans apart from running costs. They won't arrive overnight but they will arrive eventually. Remember the machines are getting smarter, we are not...

[Elus]

I don't think we need sentient machines, or anything close to that, to completely disrupt the economy.

In fact, a recent review from September 12th found that nearly 50% of all jobs are susceptible to computerization.

I do agree that we'll need good artificial intelligence and machine learning.

Like niner, I too, am not so sure about how a transition to a universal basic income or post-work society will happen, though. It could be violent or peaceful, depending on how quickly policy changes.

Edited by Elus, 23 October 2013 - 09:08 PM.

[Breezey]

The exponentiality of the pace of advancement of technology and its use is what should be taken into account. Examples cited from the past will only show rates of advancement which are slower than what they are now. When compared to the present, the advancements give the appearance of a curve that is linear, say 2X, rather than the exponential curve 2^X. But when the period of the last 30 years is more closely examined, it can be clearly seen that technology (and its advancement) is moving ahead at a pace that is far more aggressive.

Follow the green curve closely, for 0>x>4, and it is like the red curve, a straight line. IMO pre-1900s advancement, akin to using rocks, blacksmithing and using cow bladders to store and carry water. The good old days. After x=4, the curve starts to resemble the blue one, cubic growth, 4>x>8 the curve begins to represents the post-1900 world. Tanks, planes, nuclear power, low tech pharmaceuticals and the sort. Cars, the steam engine. The exact point when the curve (green) moves past x=8, is the point when hell breaks loose. I guess its some point in time that is the average of the first personal computer being sold, and mass usage of mobile phones and the internet. I'd also like to throw in the first landing on the moon into the equation.
You can see very clearly that the green curve moved up one grid during 8>x>9. It took 0>x>8 to make that same distance. The real interesting point about this whole scenario, one, that I think our dear OP Elus is trying to get at is that, we are at a time when we are going to see such advancements, such improvements in technology that we, ourselves, will look back at our own lifetimes and think about it the way, you and me think of the colour-less television right now. The possibilities are endless, with nano-technology knocking at our doorstep and space mining becoming more and more realistic. And those who say otherwise are... not that bright To the brave new world, and beyond!

Edited by Breezey, 25 October 2013 - 05:59 AM.

[Elus]

Just a clarification on the above.

Computer performance over time can be written as f(t) = 2^(t/1.5), where t is time in years. Computer performance doubles every 18 months (1.5 years) according to this formula. So, in 30 years, computer performance doubles roughly 1 million-fold. In 45 years, that's a billion-fold increase. I think it's important to be precise about these things to really see what we're dealing with.

It's fair to say that a crisis in unemployment hinges directly on the continuation of the exponential growth of computing.

If Moore's Law breaks down and nothing replaces it, then it's likely that humans will remain employed. However, if we do develop powerful AI and software that's millions of times more powerful than the kind we have today, we're in for some serious changes.

It's also important to remember that Moore's law is not the only exponential increase we're seeing. We can look at solar cost ($/watt over time), DNA sequencing cost ($/base over time), DNA sequencing volume (bases sequenced over time), and internet traffic volume (bits over time), just to name a few. These exponentials will fuel vast increases in available resources, biological knowledge, and communication, respectively. What this really means is hard to imagine.

I don't know any objective ways to measure progress in artificial intelligence, so if anyone has some concrete, objective data on that I would be incredibly interested to hear about it! The only way the judge AI is somewhat subjectively by looking at what it's capable of doing. From what I've seen so far, however, I'm very impressed.

Edited by Elus, 25 October 2013 - 06:45 PM.

[Elus]

We're seeing machine learning being applied to computer vision. Now, the hallmark remote turing test often used to assess if the user is human or not, the textual captcha, has been beaten by a machine.

Vicarious has achieved a staggering 90% solution rate in reCAPTCHA, Google's captcha software.

Forbes: AI Startup Vicarious Claims Milestone In Quest To Build A Brain: Cracking CAPTCHA

http://vimeo.com/77431982

As a bonus, here's a neural net fantasizing about what a 2 looks like:

http://youtu.be/KuPai0ogiHk?t=2m48s

Edited by Elus, 28 October 2013 - 05:44 AM.

[Elus]

A proposal for a universal basic income in Sweden is making headlines. Whispers of the idea are beginning in other countries, too. Meanwhile, we're continuing to see shifts to automation in the food service industry.

Three articles:

NY Times: Switzerland’s Proposal to Pay People for Being Alive


An army of robot baristas could mean the end of Starbucks as we know it






The Biggest Threat To Minimum Wage Restaurant Workers Everywhere?

Edited by Elus, 13 November 2013 - 02:44 AM.

[Layberinthius]

https://en.wikipedia.org/wiki/Luddite

The Luddites were 19th-century English textile artisans who protested against newly developed labour-saving machinery from 1811 to 1817. The stocking frames, spinning frames and power looms introduced during the Industrial Revolution threatened to replace the artisans with less-skilled, low-wage labourers, leaving them without work.

Although the origin of the name Luddite (/ˈlʌd.aɪt/) is uncertain, a popular theory is that the movement was named after Ned Ludd, a youth who allegedly smashed two stocking frames in 1779, and whose name had become emblematic of machine destroyers.^[1][2][3] The name evolved into the imaginary General Ludd or King Ludd, a figure who, like Robin Hood, was reputed to live in Sherwood Forest.^[4][a]

And this:
 FrameBreaking-1812.jpg   65.76KB   11 downloads

Turns to guarding factories with this:
 url.jpeg   66.09KB   12 downloads

What a wonderful world we live in. NOT!

Edited by Layberinthius, 13 November 2013 - 11:07 AM.

[PWAIN]

The luddites were wrong in their timing. Non intelligent machines taking over menial tasks were never really going to be a threat. The real issues start when machines have intelligence around the human level. That is when they can do any job that a human can. I think we are not that far away from that level now. We'll probably see more and more jobs disappear as we approach that point.

[Breezey]

What we define as being humanly possible is always changing and therefore until some machine is able to surpass human intelligence we have nothing to really worry about. Speaking from an economic point of view, with more machines taking over human jobs, jobs that require human expertise will become more competitive and more abundant. Creative works will be more in demand and people may be willing to pay premium for some goods and services with the "Made By Man" label. I will definately my food and sex to be made by humans, preferably by women.

[lemonhead]

Have you ever been to a PTA craft fair? Some people are great at making art and hand crafts, but most are not.

How do you intend to pay for your food and sex?

[niner]

I will definately my food and sex to be made by humans, preferably by women.

What if chefbots and sexbots are better than humans...? Hmm. Decisions, decisions...


As an aside, I just had this image of a guy on a couch, watching tv. He yells to his wife "Woman! Make me some sex! And a sandwich!"

[Elus]

https://en.wikipedia.org/wiki/Luddite


What a wonderful world we live in. NOT!

I am not a luddite. I love technology. I want technology to take all human jobs so that humans can be free to explore and do whatever they want.

"Tectonic Shifts" in Employment: Information technology is reducing the need for certain jobs faster than new ones are being created.


Will technology outright replace jobs or will it simply result in a shift in which jobs will be available? I happen to think it's the former, but it could be the latter. Nevertheless, if we don't adapt to the rapid nature of this shift, hundreds of millions of people could be left without jobs.

Edited by Elus, 13 November 2013 - 09:38 PM.

[Layberinthius]

that will make you a very unpopular man, I am explorer myself, it sounds like a great idea, but they sure as hell better be proofed against EMP/Solar Flares and controllable via neural links to the brain. A twist on telecommuting.

Not everyone will be for it. For example "So let me get this straight, you want to put my brain in a jar, without sunlight, food, air, or sex, and let me live in a virtual world where I can interact with other beings the same as me, as I explore the galaxy. In return you'll take my job and fill it with a robot controlled by AI who could do it a billion times better than I can?" "No thanks, I think I'll just stay here on earth, I get a nice ego boost out of being an engineer/scientist/farmer/plumber"

Some people just prefer the good old way of doing things, even in an era with longevity, virtual reality, AI, etc.

If it did happen there would have to be a significant climactic shift in the way we do things, from food production all the way up to car manufacturing, basically the earth would need to become the Star Trek universe. A very exciting proposal!

But none the less and I hope it comes true, you will probably collapse the job market in willing astronauts, 7 billion applicants for the one job is kind of high, thats another setback in the way.

Better start building ship yards.

Edited by Layberinthius, 13 November 2013 - 10:23 PM.

[Breezey]

Paying for sex is illegal and paying for food should be illegal.

[lemonhead]

What we define as being humanly possible is always changing and therefore until some machine is able to surpass human intelligence we have nothing to really worry about. Speaking from an economic point of view, with more machines taking over human jobs, jobs that require human expertise will become more competitive and more abundant. Creative works will be more in demand and people may be willing to pay premium for some goods and services with the "Made By Man" label. I will definately my food and sex to be made by humans, preferably by women.

Paying for sex is illegal and paying for food should be illegal.

So then what does the last sentence of your first post have to do with the subject of the thread?

Also, paying for sex / sex work is not illegal in some places (e.g., Nevada).

Yes, food should be free, but most people in the U.S.A. would not agree with either of us on that point (see II Thessalonians 3:10, beloved bible passage of both 20th C socialists and present-day Tea Partiers alike; I know I'm repeating myself, but I get such a kick out of the irony). Just look at what's happening with the Farm Bill.

Edited by lemonhead, 16 November 2013 - 04:30 PM.

[capob]

Industrialization in many countries brought human pain and cultural revolution (see russia, china, india, japan). There is a fallacy I've seen in this thread and elsewhere. The notion that you personally are important, and the importance of you will be reflected in the change of reality. Following, a jobless economy where your desires are provided for by helpful robots (the absurdist wall-e movie as an example). If you have this notion, you are wrong. What has and will effect the line of development is profit and power - the holding and attainment of such.

Human-job replacement is not subject to the free market. Various fallacies regarding the free market, but here I will just note on one. We live in a global oligopoly. This is not, for important matters, a free market of small (consumer) human interests. Consequently, AI is developed with oligopolists' interests. The vision, the interests of oligopolists is something found in history. But, this only matters to the extent that natural law takes over in AI, at which point you have another fallacy put forth in this thread. The concept that we control sentient AI once it is created. But, back to minimalistic, non-sentient, non-organic intelligence.

The economics of gold throughout history is great example of psychology. Likely, none of you have read Adam Smith's The Wealth of Nations. So, shortly, gold represents arbitrary distinction from others, a form of elitism. Rarely in history do you have a benevolent dictator. The term itself is absurdly selfish. So, perhaps you will have an increasing nominal socialism followed by purges of non-essentials. Take stalinism, mix in the parent replacement and blind following in hitlerism, replace the party with the oligopoly, remove the necessity of petty laborers, and enjoy the holocaust. I used to joke/tell-people about a decade ago that I had the last job in the world - programmer.

[Elus]

Following, a jobless economy where your desires are provided for by helpful robots (the absurdist wall-e movie as an example). If you have this notion, you are wrong. What has and will effect the line of development is profit and power - the holding and attainment of such.

You are incorrect. Countless times, as we've seen throughout history, a revolution has led to the transfer of power from the elites to the common man. Wealthy people who own robots/capital have built a fortune that is only a byproduct of the society in which they live. Facebook is only possible thanks to the contributions of the thousands of scientists and mathematicians who made the internet and computers a reality.Thus, these people only have a partial right to this fortune, not a complete one. Therefore, it is reasonable to assume that the general populace will take back some portion of what is rightfully theirs, namely a minimum basic income that is only capable of existing through the toil and energy derived from robots and automation software.

Technology is a democratizing force as well, which you have not taken into account. Society is becoming more transparent. Transgressions against the people are becoming more and more apparent. Human rights violations and government corruption is being leaked and exposed on social networks and ubiquitous media. Ideas spread with the speed of light, allowing untold numbers of people to collaborate on problems.

The masses will not sit idly by while elites, who own the automation and capital, take everything from them.

Consequently, AI is developed with oligopolists' interests.

AI software, once created, can be copied and distributed. Furthermore, you assume only elites will have access to AI, which is illogical. It's a bit like saying Microsoft Word, once created, would forever be confined to the elites of society. Utterly wrong. Crowd-sourced AI and public efforts, in tandem with government and private institutions, will be required to fully realize the potential of AI.

The concept that we control sentient AI once it is created.

Non sequitur. One does not need anything even remotely close to sentient AI to replace the majority of jobs that exist today.

Rarely in history do you have a benevolent dictator.

Exactly right. That's why there won't be dictators. You can look at the Arab Spring, and the social revolution. Social media and information technology have had a profound impact. They have essentially allowed masses to coordinate and congregate resistance to corrupt governments and malevolent dictators. This trend will continue to amplify as this technology becomes cheaper and more ubiquitous.

So, perhaps you will have an increasing nominal socialism followed by purges of non-essentials.

You're assuming people will remain stupid and incapable. You're assuming they won't use technology to amplify their own intelligence and skills. However, it's plainly clear that this is an incorrect assumption. As people augment themselves in radical new ways, they may have the capacity to further contribute to society's progress if they so choose. Just like the cell phone, these augmentations will first fall into the hands of the wealthy, and then follow a steep decline in price, allowing the public to take advantage of them as well.

The key point, however, is to give people a choice about what they want to do with their lives; slaving away in some pseudo-darwinian attempt to justify their right to exist is not only sad, but completely unnecessary.

Edited by Elus, 18 November 2013 - 08:55 AM.

[Breezey]

http://live.huffingt...e34444eb70002bd\

Robo Whores.


Thailand will face unemployment.

Edited by Breezey, 19 November 2013 - 05:49 AM.

[Wu Hang]

oh come on guys. the title of this topic should be
Employment crisis: Robots, AI, & automation will take most menial human jobs


just as it has always been. We are technological species and robots and automation only reflect the growth of technology. Or do you think that robots will program robots in foreseeable future? Perhaps you go as far as wars between men and machines? How very unoriginal and passe

That won't happen in a foreseeing future because strong AI (AI in which is capable of learning by itself) hasn't been developed since the idea was introduced in the 60s.

There are more than thousands of new startups dedicated to revolutionizing AI each year, some receive enormous funding, and they aren't as successful as news articles claim they are. Industrial robots have been available 50 years ago, and we have just start to introduce them into household, majorly in Japan. However even the concept isn't mainstream yet, let alone the execution, it would take another 30 years for it to really become a problem. However when that happens, we won't be the one who worry the most because at least members in this forum has been actively searching for cognitive enhancement and believe me, we are already way beyond the mainstream in terms of experimenting new ideas/concept/product.

[Wu Hang]

Industrialization in many countries brought human pain and cultural revolution (see russia, china, india, japan). There is a fallacy I've seen in this thread and elsewhere. The notion that you personally are important, and the importance of you will be reflected in the change of reality. Following, a jobless economy where your desires are provided for by helpful robots (the absurdist wall-e movie as an example). If you have this notion, you are wrong. What has and will effect the line of development is profit and power - the holding and attainment of such.

Human-job replacement is not subject to the free market. Various fallacies regarding the free market, but here I will just note on one. We live in a global oligopoly. This is not, for important matters, a free market of small (consumer) human interests. Consequently, AI is developed with oligopolists' interests. The vision, the interests of oligopolists is something found in history. But, this only matters to the extent that natural law takes over in AI, at which point you have another fallacy put forth in this thread. The concept that we control sentient AI once it is created. But, back to minimalistic, non-sentient, non-organic intelligence.

The economics of gold throughout history is great example of psychology. Likely, none of you have read Adam Smith's The Wealth of Nations. So, shortly, gold represents arbitrary distinction from others, a form of elitism. Rarely in history do you have a benevolent dictator. The term itself is absurdly selfish. So, perhaps you will have an increasing nominal socialism followed by purges of non-essentials. Take stalinism, mix in the parent replacement and blind following in hitlerism, replace the party with the oligopoly, remove the necessity of petty laborers, and enjoy the holocaust. I used to joke/tell-people about a decade ago that I had the last job in the world - programmer.

Actually programmer's job will also be replaced pretty fast too. Simply take a look at the speed in which the programming debuggers evolve from the past decade. From pure economical perspective, I would definitely use a more time efficient robot to program/debug most of the code if possible rather than hiring a high payed programmer. As a soon to be software engineer myself, I find it a lose lose situation when the programmer is asked to complete a milestone under high pressure, and due to that fact most programmers are bound to ask for a high paycheck from the boss. This cycle continues as the boss would ask a more impossible to complete task until he outsources the jobs completely to India or China. In fact, both of them are unhappy, and that alone is the main reason why nobody should look for a high paying job in this economy (absurd? if you ever have started a business before, you know that hiring a human being is both time and energy consuming. Robots are much easier and obedient don't you think?

back to the argument, yeah sure this is no way a free market. We have never had a free market existed in this world. It's termed this way to make North American (aka freedom seekers) happy yet everyone should know better that a free market doesn't involve interference from the government, monopoly and catergorization. Again AI will replace most jobs that human do, and as a human, you and I shall provide values in a more restrict manner. This means that the later generation is forced to become smarter, more organized and more techsavvy than ever before.

It's easy to predict that programming languages will be more important than traditional languages after another 50 years. However is that really something to be scared of? Human has to evolve, or they would be thrown away just like natural selection suggests. Do I want that to happen? nope, but that's a prediction, an educated guess, a good one even.

[Elus]

That won't happen in a foreseeing future because strong AI (AI in which is capable of learning by itself) hasn't been developed since the idea was introduced in the 60s.

Google brain simulator teaches itself to recognize cats

Watch Hinton's Neural Network fantasize about and learn how to recognize written digits

First, you seem to labor under the illusion that we will need strong AI to replace menial jobs. Here, the term strong AI refers to AI that matches or exceeds human intelligence. I believe you are incorrect. Many menial jobs do not require high creativity or high intelligence. Great examples of this are cashiers and drivers. In these instances, we can find automation solutions that are plainly available with technology that exists today, namely self checkout counters and driverless cars.

Second, you seem to think that AI today isn't capable of learning. This is really a very wrong assumption and could not be further from the truth. We've seen Google's neural networks teach themselves many things, including what cats and people look like. If you go to Google+, you can search your photos for different objects and features such as trees and flowers, and Google's machine learning algorithms will be able to find many of these features even if the photos aren't labeled - the algorithms have learned what these objects look like. You can make fun of its imperfections right now, but these technologies will improve on a Moore's law trajectory.

Edited by Elus, 23 November 2013 - 09:22 PM.

[Wu Hang]

That won't happen in a foreseeing future because strong AI (AI in which is capable of learning by itself) hasn't been developed since the idea was introduced in the 60s.

Google brain simulator teaches itself to recognize cats

Watch Hinton's Neural Network fantasize about and learn how to recognize written digits

First, you seem to labor under the illusion that we will need strong AI to replace menial jobs. Here, the term strong AI refers to AI that matches or exceeds human intelligence. I believe you are incorrect. Many menial jobs do not require high creativity or high intelligence. Great examples of this are cashiers and drivers. In these instances, we can find automation solutions that are plainly available with technology that exists today, namely self checkout counters and driverless cars.

Second, you seem to think that AI today isn't capable of learning. This is really a very wrong assumption and could not be further from the truth. We've seen Google's neural networks teach themselves many things, including what cats and people look like. If you go to Google+, you can search your photos for different objects and features such as trees and flowers, and Google's machine learning algorithms will be able to find many of these features even if the photos aren't labeled - the algorithms have learned what these objects look like. You can make fun of its imperfections right now, but these technologies will improve on a Moore's law trajectory.

Actually the term strong AI only refers to AI in which can feed itself and improve its judgement through that feedback. In that case, we might have similar opinion towards this.

With that aside, learning is a strong word to be used right now, as today's AI often relies on predetermined algorithms to learn. Even if it's a recursive algorithm, in which it's nothing more than a feedback system that has been developed since the 60s, it's nothing spectacular in terms of intelligence. I should have asked you that your definition of intelligence because you confuse it with knowledge. Again google's machine learning algorithm, with all due respect, can also be learnt brefly in a undergraduate level machine learning course. They have brilliant and sophisticated approach towards this, however it's still a predetermined algorithm. The reason we know is that there are many web developers who can still cheat on the algorithm and attracting viewers to their websites. Myself included.

Alright you are correct about the fact that we don't need strong AI to replace many human menial jobs. In fact, that's what we are doing right now, replacing many jobs by using industrial machinery. This is old news my friend. However if you try to replace jobs that involved uncertainty, then it really takes a long way before getting there. Namely we know that there are more than 20k vending machines in US, however none of them has replaced convenience store, which basically sales the same thing. Automatic machinery still requires supervision of human to make it work. That's what we have right now. And even the google car argument fails because they haven't been put into massive trials.

Given a simpler example, many programming courses, as well as literature courses use auto marking software to mark. And guess what, students immediately search for methods to beat the system. In fact I have to file a remarking petition every time after the auto marking results. That's the limitation of weak AI, it doesn't adapt to the environment but it rather relies on existing algorithm and the intelligence of the developers, who at least have their jobs secured because they are required. However if strong AI exists, my collegues and I all believe that every working class will be out of their jobs forever from that point. There's no way a human can compete with a 24/7 machinery which can learn from its mistakes.