Economics 2.0... should I worry?

[Elheru Aran]

Well, perhaps it was overly pessimistic. I do think it may take more than a generation, across the entire US at least. It's probably going to take a couple more decades of increased growth of the welfare rolls before people realize 'hey this is actually socialism, it's not THAT bad, we wouldn't be able to eat otherwise'.

That's if their quality of education doesn't go down the toilet first...

[Purple]

Alternatively the world could be headed for a future where humans are few and held as exotic pets by AI overlords. That's fun too.

[Dartzap]

Surely these worries are just al reaction to normal technology progression? The Luddites were saying similar things two centuries ago, after all. I don't think many of us are overly concerned with Spinning Jennies nowadays, though.

[Esquire]

My own personal take is that modern worries are of the same class as the Luddites', but we've got much more reason for it as AI really could replace humans in most tasks... assuming it's possible, economical, implemented, and not simply unplugged by the billions true AI would impoverish, which are all huge assumptions.

[Simon_Jester]

Machines threatened to replace human labor but not human thought, so there was always plenty for people to do because there's always something you can find to think about if time and leisure and resources are available.

AI has the potential to replace thought- and I don't just mean sentient computers, but also sub-sentient systems that turn out to be able to do a job that used to require intelligence. Such as automatically driving a car, or automatically finding the cheapest airline tickets.

At which point, while we can think of jobs people could still be doing, it is damnably hard to figure out how to convince a corporate-dominated economy to pay for them.

...

Moreover, on some level the Luddites were not wrong that there was massive and unnecessary suffering and dislocation of many millions of people, associated with early industrialization. A lot of people were in fact driven out of relatively comfortable lives as farmers and craftsmen and into overcrowded mortality-sink cities that were literally killing them faster than they could bear children. Work hours increased, conditions got worse, and all the while early proto-economists were piously assuring them it was for the best in this best of all possible worlds, guided by the invisible hand.

Skepticism is fully understandable.

[His Divine Shadow]

It boils down to, in my view, to the fact that AI is most likely going to eclipse human capacity. What will we do about this?

Will we become lazy, uneducated beings being sustained by our automated, AI run infrastructure until we degrade into useless appendages and eventually go extinct and leave behind our technology to continue on its own? Sort of a dystopian culture. I don't believe a true culture like scenario could emerge with people relegated to no longer doing anything important or useful in society and not even being able to remotely compete with their machine counterparts.

Will we suppress AI and force people to do jobs in order to try and pretend we're still relevant? Basically putting a big damper on human technological advancement because of our limitations.

Will we try and keep up with the increasing capabilities of AI and machines by altering ourselves, cybernetic implants, uploads, genetic engineering, in order to remain relevant, capable and in control of our own destinies?

There's also the chance that AIs just get rid of us right off the bat.

I kinda hope for the future where we try and keep our hands on the steering wheel of our own future, but without artificially suppressing AI, that's just an evolutionary dead end. If something much better and more capable can exist, it will exist, somewhere. We can't put our heads in the sand, we need to keep up.

[K. A. Pital]

The worries of the luddites were not quite unfounded, just as the worries of those who will be forced out of their jobs at the whim of the tech supercorporations and "the market" into a world where social support for the unemployed is getting more and more scarce.

[Darth Nostril]

Ah but just imagine the delicious irony if what the AIs turn out to be best at is running the corporations. All those advocates of the free market and the invisible hand will be those cast on the scrapheap first, unable to pay the mortgages of their McMansions, begging on street corners for scraps of food from minimum wage workers who wouldn't piss on them if they were on fire.

[Purple]

That's not really how I see that going. Say that AI does advance enough that it can replace the CEOs and upper management in general. What that means is that now you have a cadre of unemployed, resentful people trained and skilled at managing large groups of people. And when you combine that with the fall of middle and lower management that would inevitably precede it you have all the necessary cadres for a ludite revolution with them as the leaders and thus new ruling class.

[Starglider]

Ah but just imagine the delicious irony if what the AIs turn out to be best at is running the corporations. All those advocates of the free market and the invisible hand will be those cast on the scrapheap first, unable to pay the mortgages of their McMansions, begging on street corners for scraps of food from minimum wage workers who wouldn't piss on them if they were on fire.

That is unlikely given current trends; the neural net focused approach is approximately replicating the evolution of complexity in animal nervous systems, so complex planning and nuanced interpersonal relationships (executive skillset is mostly the same as for politics) are quite far down the list. Decision support systems are constantly improving though, for focused questions like 'which products and channels should we focus the marketing budget on'.

[SolarpunkFan]

I guess I wasn't clear on what I was saying. What I was worried about is summed up by this passage from this PDF:

What is the role of (enhanced) human beings – or of any other sentient, biological species that
produces a technological Singularity – in all this? We are initially important to the AIs as sources
of raw material, informational patterns to be simulated and assimilated. The precise way they use
the information they extract from us “is not known. (Possibilities include the study of history
through horticulture, entertainment through live–action role–playing, revenge, and economic
forgery” (313). But whatever may be the case, once our computational surplus value has been
fully extracted, we are simply shunted aside by the AIs. In some instances, they still value us as
“sapient currency units,” stockpiling us for “future options trades in human species futures”
(210). But even this sort of utility is fairly limited. Sooner or later, we are slated for “ethnic
cleansing... You take people who you define as being of little worth, and first you herd them into
a crowded ghetto with limited resources, then you decide those resources aren't worth spending
on them, and bullets are cheaper than bread” (289).

[Simon_Jester]

There's a fairly large number of assumptions built into that paragraph about how AIs behave. The blunt answer is, in the event that something like this happens, there's not much point preparing for it because it's based on the premise of future disaster that cannot be prepared for. It's basically the technophile version of the Apocalypse and the End Times.

If the underlying assumptions about how world-dominating AI networks will behave are wrong*, then... almost anything could happen. The AI could engage in permanent benign neglect. The AI could disappear into its own simulations and wind up ignoring the outside cosmos. We could disappear into our own simulations, likewise. AIs might be so nakedly hostile and dangerous that they end up being opposed radically very early in the process before their ability to dominate human society becomes reliable, resulting in a scenario I can only term a Butlerian Jihad. Who the hell knows?

But being neurotic about this is a waste of time and mental energy. There is no specific foreseeable course of action that leads from 'here' to 'there' or for that matter from 'here' to 'not there.' And whereas more mundane fears of world-ending disaster (e.g. nuclear war) could happen at any time, an AI takeover of the world is almost certain to NOT happen for decades...
___________________

*Gee, what could possibly go wrong with us trying to predict the actions of massively superhuman and inhuman intelligences? It's not as if there is some kind of... singularity... beyond which you can't reasonably predict future trends!

[SolarpunkFan]

Snip.

True, I was being rather silly with that fear.