A less optimistic view


#1

Considering the superstar model applied to AI, I would like to make some rather sceptical questions based on ideas of Nick Bostrom and Martin Ford.

First of all, the AI process will take time and will advance by very large steps. Nick Bostrom explains that first we will achieve the inteligence of a mouse, then a chimpance, then humans and after all this it will continue going further.

Then, people like journalist Martin Ford ask some really interesting questions. Maybe machines would not only be superstars in the economy. What if machines take the place of workers, self employed and the whole productive economy? What if machines throw humans out of markets? How would we distribute the income if there is not ability that markets value?

I mean, if AI is considerably superior, if machines are the superstars in a distribution of superstars, it is possible that there is no need for human employees in a situation where machines can do everything by themselves. Or maybe, this technology improvement will lead us to new works that we can´t even imagine, or will give us access to a economy with more leisure and more income equality. Actually, we don´t know.

Nick Bostrom goes further in the questions: How will power be distributed? How will be able to control it?

About this topic, TED has two interesting conferences of Nick Bostrom (philosopher of Oxford) and Martin Ford (journalist and futurist). They are really interesting. The links are:



#2

As Nesher said, if AI is considerably superior, then it is possible that all (or almost all) robots are superstars. This would lead to the implications of the model, which would mean that there is great inequality between humans and robots.

As we know, the superstar model makes some assumptions, and it is useful for explaining many issues, from large differences in income to immigration. However, I believe the issue of AI goes way beyond economics. If such a superior “beings” come to be (much more intelligent, productive, maybe stronger), perhaps even the notions of “income”, “consumption”, “utility” are no longer valid and much of economic theory will need to be reevaluated, since the assumption of a rational , utility maximizing agent may not apply to these new individuals.


#3

Raul, I totally agree with your point of view. The implications of AI on social sciences will be huge. Things will have to be restructured before being able to keep up with technology evolving at such a fast rate. I’ve got a great recommendation of someone living the implications of AI in his life.

Deep Thinking: Where Machine Intelligence Ends and Human Creativity Begins


#4

Hello Nesher, I also believe that in the case where robots are superior in the sense that the economy does not need human employees and taking it to an extreme idea where they even are more innovative then humans, there would have to be a separation in the variables of humans and superstars. In the end this could lead to what you say to new jobs or technologies developed by AI or humans that are useful to humans because of the necessity of any type of agent (AI or humans) to fill these jobs.


#5

I don’t think that all the jobs for the humans will disappear, independently of how great the robots are, there always be jobs that would require the human action, thinking or feeling. I think that the main concern is creating an environment in which this technological change can be good for all the society, like Martin Ford said in his video, we have the power to create the perfect incentives to promote high skilled jobs. One way is through bonuses that reward education, innovation and improvement; second, we also have to give the tools to the students so they can be prepare for the future (like now, that the english classes are mandatory we should give mechatronics/robotics classes as complementary of all the other subjects). Also this is an opportunity to create new markets in which the human participation is involved, like recreation activities since in the future there will be more leisure that can be use.


#6

Interesting paper on the topic of AI and economical rationality.

Assuming scientist succeed on creating machines that are perfectly rational (in the economic sense), the authors ask how to design the rules of interaction among these agents.

The paper is interesting, given that neoclassical economics begins by assuming a perfectly rational individual.


#7

I think there will be a natural obstacle that would bring significant resistant to the total replacement of human work and which presence would arrive way sooner than the AI world domination, and that would be the market itself. It is not hard to see that one of the principal reasons of the constant need of innovation in robotics is increasing firms’s productivity in order to sell more at cheaper costs, a situation seen mainly in high tech sectors of spetialized labor. But, what happens when this becomes a generalized pattern in the economies? What happens when people start loosing their jobs? You can keep improving your production all you want by intorducing new robots and technology, but, at the end of the day, you still need real people with real jobs to buy yout product.

To this, add the social problematics it would produce, especially in countries with several population of low skilled labor. I mean, taxi drivers in Mexico organized a tough, near to savage, oposition to the implementation of UBER, although it wasn’t a huge technology revolution that would cut off their jobs just like that, we´re just talking about a new competitor.

In other entries, it has been discused the introduction of UBI as a relieve to that issue, wich certainly represents a matter of thought, but I still don’t know. I mean, do humans really want to step aside and leave the leadership to machines? I would like to believe that isn’t the case, but now I’m getting into filosophal matter, so I’m going to stop there.

By all this, I think that, one way or another, technology and robotics would drift towards enhacing labor’s productivity rather than displacing it, with a very painful process in the middle, especially for the low skilled labor.

About that last thing, I found an interesting article where they explain that “two-thirds of Americans believe robots will soon perform most of the work done by humans but 80% also believe their jobs will be unaffected”. I don’t know what’s going on in the mind of that people. How to be so confidet of that?

A huge problem with low skilled labor is that you don’t even need huge technological revolution to replace it. Take a look at the cinemas or at a bank. Now you can buy your tickets or make deposits in a machine or even on your phone, where one of the principal reasons to keep counter attendants is that people are still kind of scared of using those systems, at least in countries like Mexico, but I’m guessing it wont last long until that changes.

Anyway, the article makes some points about the pattern that technology and robotics will take, regulation, affected groups and other interesting stuff. I leave the link below in case somebody find this topics interesting.


#8

Hello everyone, thanks @JuanCarlosRojas for sharing that article and @Nesher for the videos , they raised some questions on me, which I’d like to share with you and read what you think.

Let me start, recalling that losing our jobs to automation is a concern that has been present in human history for several time. Since the industrial revolution with the steam machine vs textile workers, to XX century and the invention of computers, and the result we have observed is that labour has redefined itself and adapted through time. The question we should ask then is how is this time different from the past?

I, personally, think that the major difference now is that we could reach a point in which machines will be capable of, consciously, creating new machines and improve them. That would be an unprecedented situation and I can’t predict how society will react to this scenario.

But, in the process of getting to that point, I think labour will find a way to adapt to change, maybe the loss on low-skilled job will compensate with the increased demand for other type of job, like non-routine ones as the article I share below refers, let me also quote a part of it:

“Focusing only on what is lost misses “a central economic mechanism by which automation affects the demand for labour”, notes Mr Autor: that it raises the value of the tasks that can be done only by humans. Ultimately, he says, those worried that automation will cause mass unemployment are succumbing to what economists call the “lump of labour” fallacy. “This notion that there’s only a finite amount of work to do, and therefore that if you automate some of it there’s less for people to do, is just totally wrong,” he says. Those sounding warnings about technological unemployment “basically ignore the issue of the economic response to automation”, says Mr Bessen.”

I recommend reading it because it weighs the pessimistic and optimistic view about the topic, and can contribute to the debate.

So, I’d like to hear from you, and ask you: Is this time different?

https://www.economist.com/news/special-report/21700758-will-smarter-machines-cause-mass-unemployment-automation-and-anxiety


#9

Hello everybody,
While I was reading an article I remembered about this discussion and I thought it would be interesting to share it with everybody.

I beleive this is an example of how robots can become superstars for some jobs very soon because of the type of work it takes to complete the tasks given, but as it was mentioned earlier the intelligence of the robots or the task they will be able to acheive could develop slowly and in this time people could have the chance to move to other sector that are more productive to humans. Also this is complemented by the fact that population of Japan is decreasing and it is in this country where investment for robotics is the greatest.

Here is the link to the article, I hope you enjoy it.
https://www.economist.com/news/business/21731677-around-5000-nursing-care-homes-across-country-are-testing-robots-japan-embracing


#10

I would like to add the case that is what would happen with the workers that are going to be fired for all these new technologies that are been created, and arrive at this article in which Bill Gates argues that these new robots should pay the same taxes as they are taken, but I think this is not the way to manage this situation because if we see the things from the production point of view, this new technologies will improve the production,reducing costs and possibly a reduction in prices that would improve the welfare of consumers.

What do you think?