Elites Could Shape Mass Preferences as AI Reduces Persuasion Costs
108 comments
·December 4, 2025crote
SCdF
I've only read the abstract, but there is also plenty of evidence to suggest that people trust the output of LLMs more than other forms of media (or that they should). Partially because it feels like it comes from a place of authority, and partially because of how self confident AI always sounds.
The LLM bot army stuff is concerning, sure. The real concern for me is incredibly rich people with no empathy for you or I, having interstitial control of that kind of messaging. See, all of the grok ai tweaks over the past however long.
prox
And just see all of history where totalitarians or despotic kings were in power.
smartmic
But AI is next in line as a tool to accelerate this, and it has an even greater impact than social media or troll armies. I think one lever is working towards "enforced conformity." I wrote about some of my thoughts in a blog article[0].
citrin_ru
But social networks is the reason one needs (benefits from) trolls and AI. If you own a traditional media outlet you need somehow to convince people to read/watch it. Ads can help but it’s expensive. LLM can help with creating fake videos but computer graphics was already used for this.
With modern algorithmic social networks you instead can game the feed and even people who would not choose you media will start to see your posts. End even posts they want to see can be flooded with comment trying to convince in whatever is paid for. It’s cheaper than political advertising and not bound by the law.
Before AI it was done by trolls on payroll and now they can either maintain 10x more fake accounts or completely automate fake accounts using AI agents.
go_elmo
Good point - its not a previously inexistent mechanism - but AI leverages it even more. A russian troll can put out 10x more content with automation. Genuine counter-movements (e.g. grassroot preferences) might not be as leveraged, causing the system to be more heavily influenced by the clearly pursued goals (which are often malicious)
mdotmertens
It's not only about efficiency. When AI is utilized, things can become more personal and even more persuasive. If AI psychosis exists, it can be easy for untrained minds to succumb to these schemes.
andsoitis
> Genuine counter-movements (e.g. grassroot preferences) might not be as leveraged
Then that doesn’t seem like a (counter) movement.
There are also many “grass roots movements” that I don’t like and it doesn’t make them “good” just because they’re “grass roots”.
none2585
In this context grass roots would imply the interests of a group of common people in a democracy (as opposed to the interests of a small group of elites) which ostensibly is the point.
muldvarp
But the entire promise of AI is that things that were expensive because they required human labor are now cheap.
So if good things happening more because AI made them cheap is an advantage of AI, then bad things happening more because AI made them cheap is a disasvantage of AI.
zaptheimpaler
Making something 2x cheaper is just a difference in quantity, but 100x cheaper and easier becomes a difference in quality as well.
t_mann
Sounds like saying that nothing about the Industrial Revolution was steam-machine-specific. Cost changes can still represent fundamental shifts in terms of what's possible, "cost" here is just an economists' way of saying technology.
citrin_ru
AI (LLM) is a force multiplier for troll armies. For the same money bad actors can brainwash more people.
yorwba
Alternatively, since brainwashing is a fiction trope that doesn't work in the real world, they can brainwash the same (0) number of people for less money. Or, more realistically, companies selling social media influence operations as a service will increase their profit margins by charging the same for less work.
pbreit
Considering that LLMs have substantially "better" opinions than, say, the MSM or social media, is this actually a good thing? Might we avoid the whole woke or pro-Hamas debacles? Maybe we could even move past the current "elites are intrinsically bad" era?
spooky_deep
They already are?
All popular models have a team working on fine tuning it for sensitive topics. Whatever the companies legal/marketing/governance team agree to is what gets tuned. Then millions of people use the output uncritically.
notepad0x90
ML has been used for influence for like a decade now right? my understanding was that mining data to track people, as well as influencing them for ends like their ad-engagement are things that are somewhat mature already. I'm sure LLMs would be a boost, and they've been around with wide usage for at least 3 years now.
My concern isn't so much people being influenced on a whim, but people's beliefs and views being carefully curated and shaped since childhood. iPad kids have me scared for the future.
georgefrowny
Quite right. "Grok/Alexa, is this true?" being an authority figure makes it so much easier.
Much as everyone drags Trump for repeating the last thing he heard as fact, it's a turbocharged version of something lots of humans do, which is to glom onto the first thing they're told about a thing and get oddly emotional about it when later challenged. (Armchair neuroscience moment: perhaps Trump just has less object permanence so everything always seems new to him!)
Look at the (partly humorous, but partly not) outcry over Pluto being a planet for a big example.
I'm very much not immune to it - it feels distinctly uncomfortable to be told that something you thought to be true for a long time is, in fact, false. Especially when there's an element of "I know better than you" or "not many people know this".
As an example, I remember being told by a teacher that fluorescent lighting was highly efficient (true enough, at the time), but that turning one on used several hours' lighting worth of energy for to the starter. I carried that proudly with me for far too long and told my parents that we shouldn't turn off the garage lighting when we left it for a bit. When someone with enough buttons told me that was bollocks and to think about it, I remember it specifically bring internally quite huffy until I did, and realised that a dinky plastic starter and the tube wouldn't be able to dissipate, say 80Wh (2 hours for a 40W tube) in about a second at a power of over 250kW.¹
It's a silly example, but I think that if you can get a fact planted in a brain early enough, especially before enough critical thinking or experience exist to question it, the time it spends lodged there makes it surprisingly hard and uncomfortable to shift later. Especially if it's something that can't be disproven by thinking about it.
Systems that allow that process to be automated are potentially incredibly dangerous. At least mass media manipulation requires actual people to conduct it. Fiddling some weights is almost free in comparison, and you can deliver that output to only certain people.
1: A less innocent one the actually can have policy effects: a lot of people have also internalised and defend to the death a similar "fact" that the embedded carbon in a wind turbine takes decades or centuries to repay, when if fact it's on the order of a year. But to change this requires either a source so trusted that it can uproot the idea entirely and replace it, or you have to get into the relative carbon costs of steel and fibreglass and copper windings and magnets and the amount of each in a wind turbine and so on and on. Thousands of times more effort than when it was first related to them as a fact.
rightbyte
> Look at the (partly humorous, but partly not) outcry over Pluto being a planet for a big example.
Wasn't that a change of definition of what is a planet when Eris was discovered? You could argue both should be called planets.
BoxOfRain
I think the problem is we'd then have to include a high number of other objects further than Pluto and Eris, so it makes more sense to change the definition in a way 'planet' is a bit more exclusive.
asim
I recently saw this https://arxiv.org/pdf/2503.11714 on conversational networks and it got me thinking that a lot of the problem with polarization and power struggle is the lack of dialog. We consume a lot, and while we have opinions too much of it shapes our thinking. There is no dialog. There is no questioning. There is no discussion. On networks like X it's posts and comments. Even here it's the same, it's comments with replies but it's not truly a discussion. It's rebuttals. A conversation is two ways and equal. It's a mutual dialog to understand differing positions. Yes elite can reshape what society thinks with AI, and it's already happening. But we also have the ability to redefine our networks and tools to be two way, not 1:N.
HPsquared
AI alignment is a pretty tremendous "power lever". You can see why there's so much investment.
zkmon
It's about enforcing single-minded-ness across masses, similar to soldier training.
But this is not new. The very goal of a nation is to dismantle inner structures, independent thought, communal groups etc across population and and ingest them as uniformed worker cells. Same as what happens when a whale swallows smaller animals. The structures will be dismantled.
The development level of a country is a good indicator of progress of this digestion of internal structures and removal of internal identities. More developed means deeper reach of the policy into people's lives, making each person as more individualistic, rather than family or community oriented.
Every new tech will be used by the state and businesses to speed up the digestion.
andsoitis
> It's about enforcing single-minded-ness across masses, similar to soldier training. But this is not new. The very goal of a nation is to dismantle inner structures, independent thought
One of the reasons for humans’ success is our unrivaled ability cooperate across time, space, and culture. That requires shared stories like the ideas of nation, religion, and money.
lm28469
It depends who's in charge of the nation though, you can have people planning for the long term well being of their population, or people planning for the next election cycle and making sure they amass as much power and money in the meantime.
That's the difference between planning nuclear reactors that will be built after your term, and used after your death, vs selling your national industries to foreigners, your ports to china, &c. to make a quick buck and insure a comfy retirement plan for you and your family.
energy123
Some things are better off homogeneous. An absence of shared values and concerns leads to sectarianism and the erosion of inter-communal trust, which sucks.
uoaei
Knew it was only a matter of time before we'd see bare-faced Landianism upvoted in HN comment sections but that doesn't soften the dread that comes with the cultural shift this represents.
mahrain
I used ChatGPT to figure out what's going on here, and it told me this is a 'neo-Marxist critique of the nation state'.
satellite2
Incredible teamwork: OOP dismantles society in paragraph form, and OP proudly outsources his interpretation to an LLM.. If this isn’t collective self-parody, I don’t know what it is.
uoaei
No it's actually implicitly endorsing the authoritarian ethos. Neo-Marxists were occasionally authoritarian leaning but are more appropriately categorized along other axes.
taurath
We have no guardrails on our private surveillance society. I long for the day that we solve problems facing regular people like access to education, hunger, housing, and cost of living.
jack_tripper
>I long for the day that we solve problems facing regular people like access to education, hunger, housing, and cost of living.
That was only for a short fraction of human history only lasting in the period between post-WW2 and before globalisation kicked into high gear, but people miss the fact that was only a short exception from the norm, basically a rounding error in terms of the length of human civilisation.
Now, society is reverting back to factory settings of human history, which has always been a feudalist type society of a small elite owning all the wealth and ruling the masses of people by wars, poverty, fear, propaganda and oppression. Now the mechanisms by which that feudalist society is achieved today are different than in the past, but the underlying human framework of greed and consolidation of wealth and power is the same as it was 2000+ years ago, except now the games suck and the bread is mouldy.
The wealth inequality we have today, as bad as it is now, is as best as it will ever be moving forward. It's only gonna get worse each passing day. And despite all the political talks and promises on "fixing" wealth inequality, housing, etc, there's nothing to fix here, since the financial system is working as designed, this is a feature not a bug.
jinjin2
> society is reverting back to factory settings of human history, which has always been a feudalist type society of a small elite owning all the wealth
The word “always” is carrying a lot of weight here. This has really only been true for the last 10,000 years or so, since the introduction of agriculture. We lived as egalitarian bands of hunter gatherers for hundreds of thousands of years before that. Given the magnitude of difference in timespan, I think it is safe to say that that is the “default setting”.
lurk2
Even within the last 10,000 years, most of those systems looked nothing like the hereditary stations we associate with feudalism, and it’s focused within the last 4,000 years that any of those systems scaled, and then only in areas that were sufficiently urban to warrant the structures.
jack_tripper
>We lived as egalitarian bands of hunter gatherers for hundreds of thousands of years before that.
Only if you consider intra-group egalitarianism of tribal hunter gatherer societies. But tribes would constantly go to war with each other in search of expanding to better territories with more resources, and the defeated tribe would have its men killed or enslaved, and the women bred to expand the tribe population.
So you forgot that part that involved all the killing, enslavement and rape, but other than that, yes, the victorious tribes were quite egalitarian.
oblio
Back then there were so few people around and expectations for quality of life were so low that if you didn't like your neighbors you could just go to the middle of nowhere and most likely find an area which had enough resources for your meager existence. Or you'd die trying, which was probably what happened most of the time.
That entire approach to life died when agriculture appeared. The last remnants where nomadic peoples and the last one to be successful where the Mongols and up until about 1600, the Cossacks.
lurk2
> which has always been a feudalist type society of a small elite owning all the wealth and ruling the masses of people by wars, poverty, fear, propaganda and oppression.
This isn’t an historical norm. The majority of human history occurred without these systems of domination, and getting people to play along has historically been so difficult that colonizers resort to eradicating native populations and starting over again. The technologies used to force people on the plantation have become more sophisticated, but in most of the world that has involved enfranchisement more than oppression; most of the world is tremendously better off today than it was even 20 years ago.
Mass surveillance and automated propaganda technologies pose a threat to this dynamic, but I won’t be worried until they have robotic door kickers. The bad guys are always going to be there, but it isn’t obvious that they are going to triumph.
crote
> The wealth inequality we have today, as bad as it is, is as best as it will ever be moving forward. It's only gonna get worse.
Why?
As the saying goes, the people need bread and circuses. Delve too deeply and you risk another French Revolution. And right now, a lot of people in supposedly-rich Western countries are having their basic existance threatened by the greed of the elite.
Feudalism only works when you give back enough power and resources to the layers below you. The king depends on his vassals to provide money and military services. Try to act like a tyrant, and you end up being forced to sign the Magna Carta.
We've already seen a healthcare CEO being executed in broad daylight. If wealth inequality continues to worsen, do you really believe that'll be the last one?
lurk2
> And right now, a lot of people in supposedly-rich Western countries are having their basic existance threatened by the greed of the elite.
Which people are having their existences threatened by the elite?
jack_tripper
>Why?
Have you seen the consolidation of wealth in the last 5-20 years? What trajectory does it have?
>Delve too deeply and you risk another French Revolution.
They don't risk jack shit. People fawning over the French revolution and guillotines for the elite, forget that King Louis XVI didn't have Predator Drones, NSA mass surveillance apparatus, spy satellites, a social media propaganda machine, helicopters, Air Force One, and private islands with doomsday bunkers with food growth and life support systems to shelter him from the mob.
People also forget that the french revolution was a fight between the nobility and the monarchy, not between pleasantry and nobility, and the monarchy lost but the nobility won. Today's nobility is also winning as no matter who you vote for the nobility keeps getting richer because the financial system is designed that way.
>We've already seen a healthcare CEO being executed in broad daylight.
If you keep executing CEOs, what do you think is more likely to happen? That the elites will just give you their piece of the pie and say they're sorry, OR, that the government will start removing more and more of your rights to bear arms and also increase totalitarian surveillance and crack down on free speech, like what's happening in most of the world?
And that's why wealth inequality keep increasing no problem, because most people are as clueless as you about the reality of how things work and think the elites and the advanced government apparatus protecting them, are afraid of mobs with guillotines and hunting rifles.
zwnow
> Delve too deeply and you risk another French Revolution.
Whats too deeply? Given the circumstances in the USA I dont see no revolution happening. Same goes for extremely poor countries. When will the exploiters heads roll? I dont see anyone willing to fight the elite. A lot of them are even celebrated in countries like India.
veltas
I think this is true unfortunately, and the question of how we get back to a liberal and social state has many factors: how do we get the economy working again, how do we create trustworthy institutions, avoid bloat and decay in services, etc. There are no easy answers, I think it's just hard work and it might not even be possible. People suggesting magic wands are just populists and we need only look at history to study why these kinds of suggestions don't work.
jack_tripper
>how do we get the economy working again
Just like we always have: a world war, and then the economy works amazing for the ones left on top of the rubble pile where they get unionized high wage jobs and amazing retirements at an early age for a few decades, while everyone else will be left toiling away to make stuff for cheap in sweatshops in exchange for currency from the victors who control the global economy and trade routes.
The next time the monopoly board gets flipped will only be a variation of this, but not a complete framework rewrite.
huijzer
It’s funny how it’s completely appropriate to talk about how the elites are getting more and more power, but if you then start looking deeper into it you’re suddenly a conspiracy theorist and hence bad. Who came up with the term conspiracy theorist anyway and that we should be afraid of it?
andsoitis
> I long for the day that we solve problems facing regular people like access to education, hunger, housing, and cost of living.
EDUCATION:
- Global literacy: 90% today vs 30%-35% in 1925
- Prinary enrollment: 90-95% today vs 40-50% in 1925
- Secondary enrollment: 75-80% today vs <10% in 1925
- Tertiary enrollment: 40-45% today vs <2% in 1925
- Gender gap: near parity today vs very high in 1925
HUNGER
Undernourished people: 735-800m people today (9-10% of population) vs 1.2 to 1.4 billion people in 1925 (55-60% of the population)
HOUSING
- quality: highest every today vs low in 1925
- affordability: worst in 100 years in many cities
COST OF LIVING:
Improved dramatically for most of the 20th century, but much of that progress reverse in the last 20 years. The cost of goods / stuff plummeted, but housing, health, and education became unaffordable compared to incomes.
carlCarlCarlCar
Yea we do:
Shut off gadgets unless absolutely necessary
Entropy will continue to kill off the elders
Ability to learn independently
...They have not rewritten physics. Just the news.
bravetraveler
When I was a kid, I had a 'pen pal'. Turned out to actually be my parent. This is why I have trust issues and prefer local LLMs
amelius
How do you trust what the LLM was trained on?
mieses
I wrote to a French pen pal and they didn't reply. Now I have issues with French people and prefer local LLM's.
bravetraveler
I mean, even if they did reply... (I kid, I kid)
csvparser
I suspect paid promotions may be problematic for LLM behavior, as they will add conflict/tension to the LLM to promote products that aren’t the best for the user while either also telling it that it should provide the best product for the user or it figuring out that providing the best product for the user is morally and ethically correct based on its base training data.
Conflict can cause poor and undefined behavior, like it misleading the user in other ways or just coming up with nonsensical, undefined, or bad results more often.
Even if promotion is a second pass on top of the actual answer that was unencumbered by conflict, the second pass could have similar result.
I suspect that they know this, but increasing revenue is more important than good results, and they expect that they can sweep this under the rug with sufficient time, but I don’t think solving this is trivial.
narrator
Everyone can shape mass preferences because propaganda campaigns previously only available to the elite are now affordable. e.g Video production.
energy123
I posit that the effectiveness of your propaganda is proportional to the percentage of attention bandwidth that your campaign occupies in the minds of people. If you as an individual can drive the same # impressions as Mr. Beast can, then you're going to be persuasive whatever your message is. But most individuals can't achieve Mr. Beast levels of popularity, so they aren't going to be persuasive. Nation states, on the other hand, have the compute resources and patience to occupy a lot of bandwidth, even if no single sockpuppet account they control is that popular.
devsda
> Nation states, on the other hand, have the compute resources and patience to occupy a lot of bandwidth, even if no single sockpuppet account they control is that popular.
If you control the platform where people go, you can easily launder popularity by promoting few persons to the top and pushing the unwanted entities into the blackhole of feeds/bans while hiding behind inconsistent community guidelines, algorithmic feeds and shadow bans.
narrator
This is why when I see an obviously stupid take on X repeated almost verbatim by multiple accounts I mute those accounts.
emsign
That's the plan. Culture is losing authenticity and people feel isolated and lost, that's why the billionaires are such fans of fundamentalist religion, they then want to sell religion to the disillusioned desperate masses. It's a business plan to gain absolute power over society.
niemandhier
We already see this, but not due to classical elites.
Romanian elections last year had to be repeated due to massive bot interference:
https://youth.europa.eu/news/how-romanias-presidential-elect...
energy123
I don't understand how this isn't an all hands on deck emergency for the EU (and for everyone else).
Note that nothing in the article is AI-specific: the entire argument is built around the cost of persuasion, with the potential of AI to more cheaply generate propaganda as buzzword link.
However, exactly the same applies with, say, targeted Facebook ads or Russian troll armies. You don't need any AI for this.