The Future of AI and Techno-Fascism

The Future of AI and Techno-Fascism


47%. That is the percentage of jobs that
are can be done by automations in America within our lifetime. Do you know what the
proportion is for developing nations? Two thirds, or currently about 1.6-3.2 billion
jobs in the global south if my back of the envelope calculations are correct. And it
will only grow in the future. Can you imagine what that would do to the
world? A vast sea of unemployed people, trying to make a living in a world ravaged by droughts,
floods and famines from climate change. When crop failures force people from the rural
areas into the cities, will there even be enough jobs for all of us? Mass, and I mean
massive, migrations will make this even worse. And if we look at our history, when capitalism
showed its cracks, who was blamed for everything? And who was given power by the capitalists? Now, let me ask you something. What do you
do for a living? Maybe you’re a software engineer, and you
fix bugs, or do requirement analysis or design database, or program stuff. Or maybe you’re
an Uber driver, and you drive drunk people at night and clean their puke in the morning.
Or maybe you’re a customer service representative, and you get yelled at by customers all day
long. Or maybe you’re an executive, licking the boots of capitalists and just love being
a parasite. Or maybe you’re a civil engineer, or a truck driver, or a cashier, or a secretary,
or a phone sales representative, accountant, truck driver, paralegal, fast food worker,
middle managers and I can go on and on and om Do you know what they all have in common?
All of them can be automated, maybe not tomorrow, maybe not in a decade, but some much sooner
than others. It will be a slow process, but it will be
steady. You’ll see some productivity gain here, some job loss there. Some record breaking
profit here, some wages stagnation there. At first, unemployment won’t rise that much
because there will be precarious jobs, just like how today we have Uber to fill in the
gap of unemployment. But eventually, those too will be automated, just like what Uber
is planning to do. And, as usual in capitalism, it’s the working class that will get screwed
over the most. But maybe you believe the workers will stand
together against this and join together in solidarity with all of the oppressed classes.
And yeah, people will most certainly try, but I’ll argue later that AI will make it
extremely difficult for us to unite into one single cohesive collective because guess who
will support the capitalists? Do you think we’re ready for this? If you think so, pfft I got some bad news
for you. We’re getting closer and closer to [music] This is Starcraft 2, the best game ever made
and if you don’t think so, well you’re valid and you know I’m pretty sure you have your
reasons and that’s fine So Starcraft 2 is the sequel to the classic
1998 game starcraft. In both games, your main goal is to eliminate other players by gathering
resources, building armies, attacking your opponents when the time is right, and defending
your own base against attacks from your enemies. And unlike chess or go, you have to do everything
in real time instead of waiting for your turn. You have to scout your enemy and build your
army at the same time. You have to build your bases while defending them from attacks simultaneously.
You have to manage each units carefully to minimize damage taken while maximizing damage
to your opponent’s units. It’s a really difficult game to master because there are
so many variables that have to be taken care of, while you yourself only have limited information
of the map and your opponents. You literally have to make around 3-4 decisions per second
based off of the limited information that you have and you also need to be able to infer
your opponents intention just by gleaning at their unit composition and position if
you want to be any good at it. So why am I talking about Starcraft 2? Well,
recently Google’s AI research company, DeepMind released an AI that beat pro-level Starcraft
2 players. Not just run-of-the-mill pro players either, one of them is one of the best Starcraft
2 players in the world. And it wasn’t even close, the AI beat the pro-players 10-1. I
mean they won one game, but only after the AI was made weaker. This is fucking mind-boggling to me. Starcraft
2 is an incredibly deep and complex game. Like it is several orders of magnitude more
complex than go, and go is several orders of magnitude more complex than chess. I mean
the AI would have to be able to “read” the players’ intention to know what they’re
planning to do. It would have to be able to react appropriately to situations it has never
seen before. And most importantly, It would have to be able to form complex plans while
at the same time “improvise” when things don’t go as planned. You know, like how
people do in real life? You see where I’m going with this? Okay wait I guess I have to explain why I
put “read” and “improvise” in quotation. These algorithms are not human, and we should
not think of them as human-like in any way. To say they’re reading someone and therefore
they think something is what they need to do, is not really how they work. The best
analogy I have is like the algorithm has seen something kind of like a situation, and it
has an intuition of sorts of what to do next, but again, that’s still humanizing the algorithm
so it’s still not 100% accurate. So here, I use the word “read” or “improvise”
because that’s what I think approximates the internal process of the algorithm the
best while still easily understood by most people. If you want to really understand what’s
going on, you would need to look at the math behind it. And then we have Elon Musk’s AI company
OpenAI. They developed an AI that reads some text and can answer questions based on the
text. So they basically developed an AI that has some sort of reading comprehension and
it sort of “understands” what is being written. Like if you give it some text, like
for example, an article about 2008 olympics, you can ask it questions like when it happened,
what the theme of the olympics was, and other basic questions like that. And the answers
aren’t just one or two word answers, it can answer in full sentences. It can also
write uhh articles, i guess, you can call it. And these fake “articles” make grammatical
sense and the content is coherent, but as a whole they are still meaningless. But remember
it’s probably only going to get better with time. And we’re still in the early stage of an
AI revolution, I mean we are still pretty far away from a full blown general AI with
a human-like intelligence. For example, one of the most important thing that an AI cannot
do so far is modeling reality like how we can imagine scenarios in our heads, but I
think we’ll eventually get there. So in the near future I don’t think it will be
too surprising to see an AI that is fluent in human languages, can form and execute human-like
complex plans, and improvise when things don’t go as planned. But here’s the thing, scientists developing
AI systems operate under capitalism, meaning the goal is profit rather than, say, i don’t
know, human development. This is really really really important so let me say it again: scientists
developing AI systems operate under capitalism, meaning the goal is profit. No matter how
safely the AI is designed, with all of the safety bells and whistles, it will, at some
point, harm humans because it is built under capitalism and capitalism exploits humans. Ok, so real real world example: you’re watching
this on Youtube. Youtube’s recommendation algorithm is specifically designed to maximize
watch time to optimize ad revenues and it doesn’t really care what the content is.
A video of some Alt-right jackass spewing crazy conspiracy bullshit that has inspired
terrorists will be as recommended as a thoughtful educational video if they both have the same
metric. Only after things go wrong that youtube will do something about it. Even then, it’s
only some of the time. the point is, Youtube’s recommendation algorithm
is designed for profit in mind and not, let’s say, human development or you know, making
society better in general, mostly because those are impossible to quantify and profit
from. And future AIs will have the same exact driving principle behind them: profit. Now before we go any further, I’m going
to tell you what I think is NOT likely to happen. Stuff like Skynet or shodan, that
is malevolent AI hellbent on destroying humanity for some nebulous reason, I think is very,
very unlikely to happen. Now some parts of the internet likes to speculate on how AI
will take over the world and enslave humankind because technological singularity will happen
or something like that, but I mean that’s just straight up fantasy. They assert AI will
self-improve infinitely and make themselves better and better. But, just like our good
ol’ friend capitalism, infinite growth doesn’t exist. So I wouldn’t worry about robot overlords
or malevolent AI if I were you. What I would worry about, however, is depressed
wages, and mass underemployment or even mass unemployment. I think that’s more realistic
because it has happened in the past during industrial revolution, but this time, it’s
way more than just a piece of technology. We have only been able to produce goods and
the means of production of those goods, but we have never been able to mass produce labor
itself. It takes at least 17 years to make a worker, but in the future, we’ll be able
to plop out robots and automation. Now some economists believe there will be
no mass unemployment in the future, which might be true. They’ve looked at the past
and saw that new technology usually brought about new jobs and employment. But I have
problems with this view: first, I would argue this time is different. Many economists talk
as if AI will complement workers’ jobs. But I don’t think that’s going to happen
because at some point AI will be able to just completely replace people. I mean if an AI
is at least as capable as your average person, but can be massively produced, way cheaper
and won’t unionize, then why choose humans at all? The second reason is that, when a new technology
is invented that increases productivity, wages usually fall and labor unrest follows. So
when automation eventually rolls around, wages will most definitely fall but this time, I
don’t think labor movements will be successful unless something changes. To see why this
is the case, we’ll first need to talk about. Okay, so I’m just going to start at the
Industrial revolution. During industrial revolution, skilled craftsmen and artisans were replaced
with machinery, and these machines were operated by low-skilled workers who would have been
apprentices a century prior. See the thing about craftsmen and artisans before industrial
revolution was that they owned their own means of production and they would hire apprentices
to eventually become master craftsmen and artisans themselves, thus spreading the trade
knowledge. But the invention of heavy machinery changed all of that. Now, the means of production
were instead owned by the capitalist class and, instead of training apprentices, they
hired low-skilled workers to operate heavy machinery. The workers then, didn’t get valuable
training beside whatever was necessary to operate the machinery and thus keeping them there
as labor for the capitalist class. Then, other skilled workers and artisans were wiped out
because they couldn’t compete with cheap goods produced by heavy machinery. This, along with pissed off workers who noticed
their wages had stagnated while productivity kept increasing, led to Luddite riots, where
they burned down textile factories and destroyed heavy machinery. And in turn, the government
executed many of them, like for example, after the passing of The Destruction of Stocking
Frames, etc. Act 1812. Now it’s important to note, however, the Luddites were not against
machines per se, but rather they wanted the profit to be shared more equitably, which
I mean, sounds reasonable enough, right? Except stuff like this kept happening over and over
again. That’s kind of Weird, right? So anyways, eventually, and I’m skipping a lot of stuff
here, workers were able to fight back and form unions, which gave the workers more bargaining
power and increased their wages. Anyways this is just an extremely brief summary of labor
in the industrial revolution. If you want a more detailed explanation, I recommend donoteat1’s
video on the history of labor. It’s really really good So when economists say “technology brought
about new jobs”, while yes that’s technically true, people had to die to make those jobs
not shit. They had to unionize. They had to fight business owners and the government to
increase their wages. But here’s the kicker: union memberships in countries where automation
will be deployed first are declining rapidly. And it looks like this trend won’t reverse
any time soon. What’s worse, in the future, if the only jobs left available are bullshit
jobs with bullshit pay, and workers will have to compete with each other harder, it will
make it even more difficult for workers to unionize And believe it or not, we’ve actually been
seeing it happening for a while now. See, wealth inequality has been growing for the
past 30 years. If you look at the data, specifically the US, this is happening because the middle-income
jobs are being obliterated and workers are more likely to fill in low-paying jobs rather
than high-paying ones. This shift in income distribution is caused by globalization and
automation. This results in people willing to undercut each other for low paying, insecure
jobs. As an article by Adam Booth puts it:
“Those replaced by new technology are not retrained and re-educated in order to give
them the skills required to keep up with this ever-accelerating treadmill of capitalism;
instead they are thrown onto the scrapheap and forced into the rapidly expanding “gig
economy” – a shadowy netherworld of bogus self-employment, insecure work, and zero-hour
contracts.” And in the future, if AI is capable of some
sort of critical thinking, many high-paying jobs will also be obliterated. What’s left
will be low-paying, customer facing service jobs, and with vast army of underemployed
and unemployed people, wages will be driven down even further because high labor supply
will insure that there will be people willing to take the lowest pay. And this is going to be worse for the global
south, yet again, because capital moves but human labor can’t. In the past, we’ve
seen the outsourcing and offshoring of manufacturing sector to developing nations because labor,
usually the most expensive cost of a business, is way cheaper here. But robots will be even
cheaper, because they don’t ask for a raise, they don’t unionize, they don’t need off
during holidays, they can run literally 24/7 and so on and so on. And because capital moves
and human labor can’t, manufacturing sector will eventually move back to rich countries
again, except with little to no job gain in those countries. This will screw over everyone,
especially developing countries yet again, because it will leave a vast amounts of people
unemployed all over the world. Imagine what that world would look like, when
a big portion of the people will have to live paycheck to paycheck if they have jobs at
all. Social mobility will be damn near impossible. And the version of UBI that is currently being
championed won’t help with that, but that’s another video. I mean, look at western countries
today, ya’ll have stagnating wages and gigantic income inequality, and everyone is damn near
pissed off all the time. Can you imagine what a much worse version of today would look like?
Well, let me tell you about. Techno-fascism. Yeah, that sounds really dramatic.
And okay, I admit this next part is mostly just me speculating on what can happen given
some, I would argue, reasonable assumptions, so you know, take everything I’m saying
with the biggest grain of salt you can find. But I think it’s important for us to recognize
how technology might help fascists in the future, so that we can fight them more effectively. Oh and also, before we continue, I’m pretty
sure you probably have a good intuition of what fascism is, but I would still recommend
you read Umberto Eco’s Ur-Fascism, I mean it’s only like 10 pages long, or if you’re
more of a visual medium type like me, ContraPoints, Philosophy Tube and Angie Speaks have all
done great videos on fascism. Also, in this section I’m going to lump in alt-right,
neo-nazi, white nationalist and the like as fascist, though in other context it’s not
exactly accurate. Fascism has been on the rise lately. Not just
in the western countries, either. I mean, I’d argue ISIS is, or was I guess, a fascist
group because they tick almost all of the boxes. I mean, they’re traditionalist, they
reject what they call western decadence, shun critical thinking, they stamp out ideological
and belief diversity, they’re obsessed with martyrdom and they’re essentially a death
cult. You also have leaders like Erdogan, Orban, Bolsonaro and Duterte, which flirt
with straight up fashy ideas and do fashy things. And then, of course, you have western
countries themselves. Like literally all of them now are plagued by fascist or fascist-adjacent
organizations. I’m not going to name names here, but you probably know what I’m talking
about. So we really should ask ourselves, why the hell is this happening now? Well, some might argue it’s a reaction to
terrorism, specifically fundamentalist Islamic terrorism. But, I mean, I used to live in
the US for like a decade at the height of the war on terror, and the fashy elements
weren’t this overt. Sure, there was a patriotic fervor going on, “I support the troops”
and all that, but there weren’t people advancing ridiculous conspiracy theories about white
genocide, or other fashy bullshit like that, at least not in the mainstream. The xenophobia
and racism back then was more let’s say subtle, under the surface, kind of condescending,
even. Nowadays though, it’s just straight up out in the open, with people being able
to say shit that would’ve gotten them ostracized just a decade ago. So what the hell happened? Well 2008 happened. Two things. First, world’s
economy crumbled apart and threw so many people into precarious economic situation. Second,
right-wing fearmongers blamed the liberal elites, which in this case was represented
by Obama, for the economic meltdown. They weren’t exactly wrong, but they then went
further and started a narrative that lump in together pretty much everyone not on the
right with the elites, especially left-wing activists, immigrants and ethnic minorities.
I mean people like Bill O’Reilly and Glenn Beck pretty much started the narrative about
how the liberal elites along with minorities and immigrants, who are all secret commies,
hate America because something something freedom And it was really really profitable. For a
time, Fox News was the most watched news in America. I remember going to a friend’s
parents house and it was always Fox News playing on TV. Other grifters then saw how profitable
it was to scaremonger and exploit the atmosphere of fear due to economic precarity. Instead
of blaming the rich and the politicians who helped them, these grifters blamed the others:
immigrants, people of color, the poor, LGBTQ+ people, secret cabal of socialists and globalists
and so on and so on. And as time goes on, and the narrative got wilder and wilder, until
eventually fascists were able to embed their ideas in the mainstream. So now here we are,
with the free market at work spreading misinformation and lies through a torrent of deceptive social
media posts. And, as long as this capitalist crisis is
not addressed, fascism will linger. On the video “the function of fascism” by Kay
and Skittles, which is really really good by the way, they said “fascist movements,
much like revolutionary movements, come into existence in times of crisis. When the wealthy
class of landlords and business owners fears the possibility of an impending worker uprising.
And/or when the monopolies that capitalism produces leads to a decline in investment
and a possible financial crash. Fascists allied themselves with these economic elites, striking
deals for their support and funding, in return for them ensuring they protect the economic
and social status of those elites. Thus fascist governments have always protected and, indeed,
championed inequality and corporate interest, while leaving the option open for state intervention
in the favor of corporate interests”. That sounds eerily familiar, right? Right now,
wealth inequality is at an all time high. And people are scared because they’re in a precarious
economic situation. And then there are opportunistic assholes spouting nativist bullshit fearmongering
about refugees, and unfortunately climate change is just going to make this worse in
the future. I mean we’ve seen this in the past, capitalism got into some sort of crisis
of its own making, fascists started to blame the others, then the capitalist class gave
them power and tried to preserve the capitalist structure. And as more and more wealth are accumulating
at the top, how likely is it, do you think, that the capitalist class will be willing
to redistribute their wealth? I mean in the past, they would rather hand the power to
fascists to smash leftist movements and rollback worker protection laws than share their power.
And this is where it gets a lot worse for almost everyone else in the future. See, right now, the technological capital
for everything is solely in the hands of the rich. From the undersea cables to millions
of servers in server farms to the software used to run the internet, they’re all owned
by them. And no matter what anarcho-primitivists tell you, we do need the internet to live
in this modern world. Right now, humans still control the internet, but more and more tasks,
mostly moderation and such, are being done by AI. And I argue that at some point in the
future, if AI is used to make decisions that run companies, AI will favor fascist groups
rather than leftist ones based on 6 assumptions: Assumption number 1. The current neoliberal
system is unstable because wealth inequality is increasing and climate change will destabilize
the world economy. If you haven’t seen the first part of this video, you should check
it out. Because over there, I talked about how climate change is going to destabilize
world’s economy. This will be on top of automation, which will replace a lot of workers, possibly
causing mass unrest. Assumption number 2. Worker/leftist movements
harm the current capitalist structure Assumption number 3. Fascism has been successfully
used to preserve capitalism in the past. See Francisco Franco in Spain and Augusto Pinochet
in Chile. Assumption number 4. AI will be designed to
maximize profit, and hence requires the preservation of capitalism
Assumption number 5. In the future, AI will be able to model the world accurately to a
certain extent. Which leads us to. Assumption number 6. AI will be used to make
many high level decisions in big companies If all of those assumption holds true in the
future, AI will then be more likely use its power to prop up fascists. This is what I’m
going to call techno-fascism, or fascism propped up not necessarily by the rich or the capitalist
class, mostly because they might not even realize the machines are doing it, but by
algorithms whose main objective function is to maximize profit. You can kinda already
see this happening today with algorithms recommending fashy contents, though for completely different
reasons. But so far, all of them have been stopped because someone, somewhere said “Hold
up, this is bad. This looks, sounds and smells like fascism. This is bad, we need to stop
it!”. Because you know, some people still have morals. But machines don’t, at least
not inherently. Now, don’t get me wrong, ideas that implicitly support fascist ideology
are widespread on social media, dominating even, but straight up fascist contents are
still removed when they pop up, that’s the way it’s supposed to works anyways. The
reality might be a little bit different. See the current fascist objectives don’t
quite line up with the current capitalists objectives. Selling hate still tends put
a lot of people off, so tech companies don’t really want to be seen cozying up with those
people. At some point in the future though, that might change. If worker movements start
hurting their bottom line, then it is only logical for them to try to quell those movements
without having to spend a lot of money. A great way to do that, if we look at history,
is to prop up fascists and weaken labor movements. And again, these decisions might not be made
by humans, because remember, a lot of them would have been replaced by machines by that
point. I mean let me put it this way, if AI ends
up controlling our means of communication, what’s stopping them from subtly manipulating
what we see or hear on the internet? “What?” you might scream at the monitor,
“that’s goddamn outlandish, ridiculous! How would the AI even do it?” Well, dear
viewer, I’m glad you asked. Do you use twitter or reddit? I mean if you’re
watching this, chances are you found me on reddit or twitter. Maybe it was on the breadtube
subreddit, or you know, maybe a generous soul decided to share this video, which if you did, I genuinely appreciate it. But who gets to choose what you see on your feed? Like there’s no way in hell a real person is doing that because there’s just too much
stuff streaming from people’s brains straight to the internet, so there has to be some sort
of algorithm that picks them for you. Let’s take twitter as an example, and let’s
say you’re following between 800-2000 people, which, from my cursory research, seems like
a reasonable number. Now, 500 million tweets are sent per day, and there are 326 million
active users on twitter. This mean on average each active user sends about 1.5 tweets per
day. So there should be around 1200-3000 tweets flooding your timeline everyday, and this
is an underestimation because you follow people precisely because they tweet stuff that you
find interesting so they probably send more than 1.5 tweets per day. Now obviously not
all of those tweets will be shown to your feed, so twitter algorithm picks them for
you. I don’t exactly know how it works, but if I were to make an educated guess, it
probably looks at engagement metric, how much interaction you have with that person, language
usage, time, and stuff like that. Now what if some pernicious god wants to change
your belief. Do you think it would be able to do it with just twitter? I mean, maybe
it starts with twitter suddenly recommending a person for you to follow, so you follow
this person. Maybe this person makes funny memes or some shit that you really really
like. You press like on all of their tweets, you retweet a couple of them that
you really identify with, and slowly this god is building your personality profile.
Next, this person retweets another person and you find their tweets really interesting,
so you follow them too. This keeps happening until most of the people you follow, you know
them through that one person. And this god would be able to change your belief just by
flooding your feed with content geared towards certain perspective. Do you know why I think this can happen? Because
it happened to me, although it was to the left and with Youtube. I mean, I wasn’t
a reactionary or a conservative, but I was a liberal. Then one day youtube recommended HBomberguy’s
video on Sherlock, I think it was, and I thought, hey this guy’s really funny, so I checked
out his other videos. One video led to another, and that’s pretty much how I got started
on leftism, a video about how Sherlock is garbage. Which is kinda funny I think. The
system used by tech companies today, however, is undirected. What’s recommended is what’s
popular, and what’s popular is recommended by the system. It’s sort of blind, only
reflecting the biases of the people who are using it and the people who designed it. In the future, though, AI might be self-directed
towards certain ideology that can help complete their main objective, which is profit. If
the AI thinks some content, any content, harms their main objective, it will simply just
not recommend them or they won’t show up in feeds. If they need to change minds, they
can start by building personality profiles, then recommend contents closest to the beliefs
of the targets, and then recommend more and more progressively fashy contents that line
up with their objective. And finally, If they can’t change someone’s mind, then they’ll
just quarantine that person within an isolated group. Actually let’s get into more detail on how
an AI might isolate groups and individuals. See, if the AI doesn’t like a person’s
post, for example, if it has some sort of anti-capitalist message, it can show that
post only to people who disagree with it, essentially making sure they’re attacked
for their view. And it won’t have to be real people either, bots will do the trick.
It will make it as if that view is unpopular in the eyes of that user, and that might cause
them to change their mind. If that doesn’t do the trick, the AI can instead isolate the
person by nudging them into an isolated social group within the network. The AI can do this
by doing the reverse, showing the posts only to people who will agree with them on a set
of extremely narrow topics all of the time. And again, it doesn’t have to be real people,
bots will also do the trick here. These groups will rarely interact with the
general public or each other, or if they do, it is specifically done to pit groups that
are normally allies against one another as a way to establish distinct and detached group
identities for their members. To make the groups seem more legitimate, bots can be used
as a sort of filler, making sure to keep advancing some narrative that can be used to perpetuate
the group. This is what I meant when the title said atomize: divide people into smaller and
smaller groups that are extremely isolated, and can’t communicate with each other or
when they do, they’re fighting one another. So in the future, if leftist groups and people
are atomized into smaller and smaller units, it will be harder to gather enough people
to organize any sort of collective actions through the internet, but it will be the reverse
for the other side, because remember, the AI’s objective is profit which requires
the preservation of capitalist structure and fascists preserve capitalist structure. It’s
logical then, to conclude that it will be much easier for fascists to organize precisely
because they will be propped up by the AI. It will also be harder to disseminate leftist
information through the internet, not because it will be removed, but rather because it
will only circulate in smaller and smaller groups, without ever coming into contact with
the general population. And again, it will be the other way around on the other side. I couldn’t fit this anywhere else, so I’m
just going to say it here. What I’ve been talking about since the beginning of this
part is essentially an extension of an article by, and I’m sorry if I mispronounce his name,
Francois Chollet’s. I think. It’s a speculative opinion, his words not mine, about how social
media will be used to control people in the future. He talked about the technical detail
on his article, and I added the political stuff. Not really related, but this person
is how I got started with deep learning in the first place because he made one of the
best, easiest to use neural network framework out there. So you know, if you want to read
it straight from the expert, check out his article. But unlike climate change, which requires
the overhaul of our society to fix, this one is totally preventable, so let’s talk about. Of course none of these are guaranteed to
happen. But seeing what happened a couple of weeks ago, we really shouldn’t leave
it to chance. We really need to try to make sure automation won’t lead us to a dark
path filled with economic insecurity and techno-fascism. Now, granted, I’m just some idiotic asshole
from the global south, but I have a couple of suggestions: First, we have to keep fighting fascism in
all of its forms, both online and offline. Support your local antifa. Deplatform fascists
whenever possible because it actually works. Expose them to the world. We need to make
sure fascism will never be an acceptable ideology anywhere on earth. Second, build strong communities. Both offline
and online. Whatever some people might say, the internet is a great way to build communities.
I know that sounds like some neoliberal bullshit, but it’s actually true. Get to know your
comrades and don’t fall into bullshit call-out culture. People, wherever they are, will tend
to form groups, but make sure those groups are not isolated. I mean ya’ll don’t have
to agree on everything but cooperation between groups is really damn important. Build strong
connections through multiple channels, just in case some of them are compromised. And
most importantly, translate whatever you have online into real world action. We have to
keep this up in the coming decades too, so people who are already experienced need to
teach how to organize and build communities to the next generation because the fascists
will do the same and we can’t be fragmented to fight them. Third, support the decentralization of technology.
See, most neural network architectures are open to the public. The reason tech companies
can do that is because the architecture is not the hard part of building an AI, gathering
huge amount of data and training it are. I’m pretty sure in the future there will be projects
that will try to build AI, but open source and with better objectives than megacorporations.
The problem is that it will need a huge amount of computation to train it, so they will probably
use distributed computing to do that. You will be able to help them by giving some of
your computing power to them when yours is not being used, like how CERN, the hadron
collider people, use distributed computing to help them run physics simulations. And
if you’re a computer scientist or a software engineer, look up this paper, the link should
be in the description. Lastly, and I think this is the most important
part, the underlying condition that caused all of this needs to be changed. I’m not
going to pretend like I have the answer to this one but right now, the least anyone can
do is to vote. That’s like the bare minimum anyone can do. Now, do I know for certain these things will
happen? No, of course not. I’m not Nostradamus or David Blaine or Jesus. With climate change,
I’m not the one who made the catastrophic predictions. Scientists have been screaming
about it for the past 2 decades. With underemployment brought about by AI, also not me. Economists
have been saying that if we don’t do anything, shit will hit the fan, and they’re been
saying it for the past 10 years. But what I’m putting forward is some sort of logical
conclusion, given that both climate change and automation will inexorably change our
world, for better or, more likely, for the worse. And as a response to that upheaval,
I think many fascist or fascist-adjacent governments will pop up all over the world. They won’t
probably call themselves fascists, but it will be some sort of nationalistic and capitalistic
authoritarianism. And this brand of fascism might wax and wane over the coming decades,
but I think it will grow more and more popular over a long period of time unless we do something
about it. Or maybe I’m just wearing doom tinted glasses
and wildly extrapolating from trends that won’t last. But do you want to risk it? Hey you got this far, Thank you for watching.
Don’t forget to like and subscribe and follow me on twitter at whatever’s on the screen
right now or in the description. If you like this video, maybe the idea of the video, maybe
not so much the execution of the video, maybe share it because I know the execution of this
video is not very good. But I would really appreciate it. What else? The next video I’m
going to do is on fascism again, I think, or UBI. So you know, subscribe I guess. Anyways
BAAIII

43 thoughts on “The Future of AI and Techno-Fascism

  1. I really like your videos, your writing and your style. The quality is really good for such a small and new channel! I hope you make more going forward

  2. hey bro just wanted to say I love your channel and I love your style, it's really smooth and flows so well it makes half an hour go by insanely quick

  3. I love your videos, I'm happy there are people like you who take the threat of AI under capitalism seriously. If you haven't already, do you plan on posting your video to any Breadtube forum like the subreddit? I would like to see this video get as much exposure as it can

  4. Nothing can grow infinitely, but assuming an AI system approximately the size of a human brain running sufficiently advanced software can be as intelligent as a human, then a superhuman intelligence wouldn't take much resources

  5. Great video. Not sure why you were so self conscious at the end (it was kinda cute tho). I love that you cite your video, looks like a lot of work. Hope you keep it up. Thank you for all you do.

  6. I know you just started putting out content, but you already deserve a ton more subscribers! People really need to listen to someone like you, someone with your views, with your perspective more than ever!
    Keep it up comrade, your work is invaluable, and very informative!

  7. … That wiki description of technofascism included "chemotherapy" under the technology it utilises.

  8. I don't think it's possible to develop self-aware AI using classical Turing systems. But if these AI algorithms are shown to have performance gains on quantum systems compared to classical, that will be a game changer.

  9. You're title card music is too loud relative to the volume I listen to your voice at in more than one video.

  10. I love how many people hbomb has turned to leftists, same story here. Really good video, keep up the god work!

  11. I have faith in humanity. I'm sure we can come up with something even worse than fascism or capitalism.

  12. Great work. I'm going to add a few extra words here to stimulate the algorithm. I'll also add the words 'cool products that I'm totally going to buy right away' just to see if that gets picked up and turned into better metrics. Seriously, thank you for making these. It's going to help a lot.

  13. If AI got good enough it'd understand that to maximise profits it needs to end neoliberalism which is destroying us all… one can always hope.

  14. If socialism is responsible for kill 100 milion people, neoliberalism may extinct the whole mankind!!

  15. Do you think most people supporting fascism even know what it is?
    The amount of disinformation that has propped up capitalism seems to keep people confused about the meanings of words. It's worse than 1984, with the environmental side, and even people that think they are going green create massive amounts of damage just in their everyday lives.
    Who speaks about cutting back their own goals in order to align with the environmental reality? Most people are still trying to achieve and accumulate wealth… I know I'm speaking from a bubble though, in a well off community of the global north.
    I think most people around here have their heads up their techno utopian asses….
    Either the privilege has prevented critical thinking, or the fear has paralyzed us. I can't tell until I talk to people, but most refuse to address any deep conversation that threatens their sense of comfort.
    I know the global south doesn't get that luxury, and am trying to raise awareness and get people to care. Ideas?
    Mostly people just unfriend me when I try to catalyze a movement towards something better. They think it threatens their well being, even though it is completely the opposite. This is where it gets Orwellian, because they have terrible rationality that circles around false beliefs. Eventually when confronted with a point they can't disprove, they just "agree to disagree" or have "their own beliefs"
    I am working on better ways to address these things with them and try to be patient, but we don't have the time to wait. We need to stop the harm we are causing…

    Anyway, I don't know why I'm ranting, I guess you just raised a lot of amazing points in that incredible video. Great work bro, and keep it up! We really need more people connecting the environment, capitalism and technology. I see too many people funnel into one focus because of the profit algorithms; don't fall into that trap!

  16. Man, this was so good. Subscribed, liked, doing that comment thing for the AI overlords. Can't wait to watch the rest of your stuff.

  17. This is one of my biggest fears. I am mainly worried about AI news and information. If anyone watched that video about the Most Profound Moment in video game history in MGS2 (Metal Gear Solid 2) by Max Derrat, you will know what I mean. That game chills me to this day.

  18. I started with contrapoints, though ian danskin is currently my spirit animal. This will be a good video to show my friend who's a bit of an Elon Musk fanboy.

  19. Thanks Ray, a great and rare video on technology and from leftwing perspective – a REAL lefting perspective that is.

    I am in Malaysia and I recently thought of the term "techno-fascism" to describe my state government's decision to implement a parking payment application for mobile smartphones across Selangor state, which links to one's credit card or bank account, whilst at the same time making it awkward to pay for parking with cash through an automated parking machine operated by the city local authority (which comes under the state government) at the park & ride car park beside a light-rail transit station near my home – i.e. it only accepts one ringgit notes and payment using this parking app.

    Whilst smartphones are very affordable these days and almost everyone from babies to octogenarians and even nonagenarians of all income groups ranging from the poorest to the richest have one, however from my own observations, most of them only know how to use their smartphones to access WhatsApp, Facebook, You Tube, Twitter and other such social media but are unable to handle a payment app which employs two-factor authentication with a back-end link to their bank account or credit card account if they have one.

    It's pretty obvious to me that the local authority is making it difficult for those who are not-savvy enough or who refuse to use the state government's smartphone payment app, to pay with cash. Whilst I would have no problem using the app, however I refuse to, especially when I have to allow it access to my bank or credit card account. Also, there has been an long tried and tested, and much simpler and more straightforward means to pay parking electronically available in Malaysia for 20 years now in the form of a credit-card sized, contactless stored-value card which I use to pay for my transit fares and for parking in buildings and shopping complexes, which can be topped up by paying cash across the counter to counter staff at petrol stations, convenience stores, at the metro train stations and so forth. However, my local authority's parking payment machine does not accept payment using this contactless payment card.

    This is what I mean by "techno-fascism" within this particular context.

    And on the much touted buzzword – "Industry 4.0" doing the rounds of the ICT industry and Malaysian government policy today, its advocates and proponents claim that its adoption will create more jobs higher up the skills chain, which will "replace production floor jobs lost from the adoption of Industry 4.0 production facilities on the factory floor" and they go even further to claim that Industry 4.0 will "increase the competitiveness" of industries,which will "enable them to expand and create even more jobs". Others even claim that the adoption of Industry 4.0 in production industries will "enable new jobs to be created" in other areas.

    From the experience of the Malaysian banking sector, the introduction of automated banking such as self-service automated teller machines, cheque deposit machines, cash deposit machines and well as Internet and mobile banking, starting since the late 1970s, has gradually seen the reduction of the number of counter clerks in banking halls to a relative handful, with the closure of many bank branches, despite claims by senior bank managers that those staff displaced would be re-depolyed elsewhere within the bank.

    I thus see the same happening in Malaysia, with as especially manufacturing industries move towards full Industry 4.0 over the next 15 years.

    Also, according to our Ministry of Higher Education, graduate employment is only 50% with employment amongst ICT graduates being even lower at 40%, which when put the other way, graduate unemployment is 50%, whilst unemployment amongst ICT graduates is 60%. Meanwhile, the graduate factories continue to crank out graduates, including engineering graduates who end up working in the gig economy as "Food Distribution Engineers" delivering food to customers' homes and offices via motor scooter, whilst many others use their cars as e-hailing taxis for a living or to earn additional after their day job.

    On the other hand, the government tells us that we need more STEM (science. technology, engineering and mathematics) graduates but for what – to work in the gig economy?

    The government is now considering a proposal by our Youth and Sports Minister that we introduce Go-Jek motorcycle taxi e-hailing service in Indonesia to Malaysia in order to provide gig-jobs for Malaysian youth.

    We leftists are not Luddites as the marketing touts of the ICT industry disparage us as being – Instead we know that when "Industry 4.0" is being touted to capitalist industries as a means for them to reduce their number of human workers, hence their labour costs, that the same capitalists are not going to hire the same number of higher-skilled workers on higher pay to manage the Industry 4.0 production systems, than the assembly floor workers which Industry 4.0 made redundant by it implementation,

    We know that only within a socialist system, will the technological advancements such as Industry 4.0 produce in abundance to serve social need, not private profit and enable the ultimate realisation of – "From each according to his ability, to each according to his need".

     

    BTW. I see you are a Linux user like myself, from you Ubuntu Linux desktop. I use the Ubuntu-based Linux Mint with the Cinnamon desktop, having given up Ubuntu after they dropped the Gnome 2 desktop environment.

    No offence intended comrade, but I found the Unity desktop more cumbersome to quickly navigate than Gnome 2 or Cinnamon.

  20. Thank you for putting things I couldnt quite formulate into clear and concise video. I study CS, and I have been deep in different fringe cyber-philosophical online groups ever since I can remember(first non-fiction book I have read was Cyberia by Douglas Rushkoff) and thats how I got to transhumanism, anarchism and post-modern philosophy. I will definitely read the whitepaper from your sources.

    Another thing you can all do: If you ever get some right wing grifter recommended by youtube or on other social media, dont ignore it, there should be an option of unfollowing(in youtube its "I dont want to see it" option hidden behind three dots next to video thumbnail or video title, then use "Tell us reason"). Dislike nazi videos if you ever get to one(but dont search them on purpose and click off immediately) , use adblockers but whitelist channels you support. Youtube especially, after bunch of debacles(Weird children videos, softcore pedophilia, nazis, adpocalypse, etc) are now doing changes to algorithm where they are trying to gauge "quality" of videos, not just by likes and retention time, but they actually sometimes show you a poll on main page about a video you watched recently. FILL THE POLL IN, RATE VIDEOS YOU AGREE WITH 5*.

    AI isnt some magical know it all and data are not some mysterious substance. They are just bunch of linear algebra and some information about where you click and how much time you spend on what. If nothing else, at least we can try to bias the algorithms towards our goals. Its still pretty weak, we still have to work with neolib tech company, but if we can convince them that left is more profitable than right, we can at least radicalize some people along the way.

Leave a Reply

Your email address will not be published. Required fields are marked *