'Slow this thing down': Sanders warns US has no clue about speed and scale of coming AI revolution

[BumRushDaShow]

Source: The Guardian

Fri 20 Feb 2026 23.26 EST
Last modified on Sat 21 Feb 2026 03.00 EST


Bernie Sanders has warned that Congress and the American public have “not a clue” about the scale and speed of the coming AI revolution, pressing for urgent policy action to “slow this thing down” as tech companies race to build ever-more powerful systems.

Speaking at Stanford University on Friday alongside congressman Ro Khanna after a series of meetings with industry leaders in California, Sanders was blunt about what he called the “most dangerous moment in the modern history of this country”. “The Congress and the American people are very unprepared for the tsunami that is coming,” he said.

Khanna, a progressive Democrat who represents Silicon Valley, shared Sanders’s concerns, warning that the country was experiencing a “new gilded age” run by tech billionaires who believe “they would have been heroic conquerors in a different era”. “That’s just not my observation,” Khanna said. “That’s what they tell me.”

Khanna and Sanders declined to specify which tech executives they met with during the senator’s visit to California, but the congressman said it was “senior leaders” at the “most prominent tech companies”. “I think it was important for both Senator Sanders to hear from tech leaders and tech leaders to hear from Senator Sanders, who represents and understands the concerns of so many working-class Americans,” Khanna said in an interview after the event.

Read more: https://www.theguardian.com/us-news/2026/feb/21/ai-revolution-bernie-sanders-warning

[William Seger]

1. What's frightening to me is how many people think AI is superior to actual intelligence.

The AI we have today only simulates what is only a subset of what actual intelligence is capable of, and occasionally produces pure nonsense, but the reason it's heading toward us like a freight train is because it does that subset cheaper than actual intelligence.

[Marie Marie]

7. I heard an interesting point about AI's self learning from all the data it is absorbing.

It is beginning to also absorb AI hallucinations and AI slop and incorporate all that misinformation into its "higher" learning. Great.

[cynical_idealist]

18. plus when AI gives answers they don't like, they poison the input to shape the output

Dear AI, how do we fix global warming?
AI: Stop burning fossil fuels.

[twodogsbarking]

2. We are all on the train but we're not sure where it is going or if it will jump the tracks.

I agree with Bernie. Also, though, handing things to the government ain't always the best decision either. Will AI determine the fate of AI? Oh, the irony.

[Jim__]

3. Unfortunately, the Trump administration has it's head up it's ass. It won't do anything.

[not fooled]

17. And its hand in the Silicon Valley moneybags

the tech oligarchs made sure to well and truly buy off Diaper Don. As well as facilitate if not make inevitable his re-installation.

[LudwigPastorius]

4. Trump will double down, not slow it down.

I foresee a huge government bailout coming for Sam Altman and OpenAI.

The Chinese must not achieve Artificial General Intelligence before the United States because of something, something, national security something.

[dickthegrouch]

5. Abundant Iniquity is my expansion of AI

It is Artificial and in no way, shape, or form intelligent.
It was programmed by mostly well meaning people, but is being trained from iniquitous news media, and illegal scraping of intellectual property.
It is not intelligent enough to discern fact from fiction. Especially when neither the government nor the press, two previously trustable resources, are putting out anything more than fiction.

Just because the word truth appears in some part of the page, does not make it so. No amount of repetition of the lie actually makes it truth, but AI systems are programmed to react to frequency, not veracity; they are as incapable of appreciating veracity as most republicans.

Intelligence requires discretion. AI will never have that, IMHO.

[SWBTATTReg]

8. Garbage in, garbage out, eh? Wise thought by you here, one that I agree w/, anyone in IT knows that.

What gets me is that who decides to what elements/knowledge to impart into an 'AI' entity being created? And of course, what will be its
primary inputs and / or outputs. So many issues to address regarding the whole scope of an artificial AI.

[area51]

13. You mentioned AI not being intelligent enough to discern fact from fiction;

that certainly decribes maga.

[Mosby]

6. I just read an article in the Atlantic

About the growth of AI, some quotes:

Jim Farley, CEO of Ford, estimated that it [AI] would eliminate "literally half of all white collar workers" in a decade.

Dario Amodei, the CEO of the AI company Anthropic, said that AI could drive unemployment up 10 to 20 percent in the next one to five years and "wipe out half of all entry level white-collar jobs".

Tell your kids to skip college and learn HVAC. Those graphic cards aren't going to cool themselves.

[SWBTATTReg]

10. This date is very optimistic. It costs lots of money to replaces the millions and billions of line of code for the AIs

and of course developing the new platforms, machinery for the AIs to reside in, perform in.

[GenThePerservering]

21. I hear you

but there is an awful lot to be said for the skilled trades! This might be motivational.

[creon]

9. AI equals GIGO

It is not intelligent. It has none of the important qualities of intelligence.
IFirst thinkgs, it i mainly clerical. It will eliiminate low level clerical jobs.

Finiancially it is bubble. There are billions of dollars od debt. Debt that cannoy be paid.

[jfz9580m]

15. I honestly think the hype is dishonest

I have given up on the reality being taken seriously so criminal liability for ai and data mining is the only real fix I can think of.

Or these jerks will easily and duplicitously game the legal system, using every crisis to consolidate their power further.

I have seen two worlds by now thanks to ai and in both EECS sucks and is dishonest but it is hard to prove as they pull of this tabloid gossip’s level understanding of human nature with strategic data flows/logistics.
Otoh it is expensive, clunky brute force with none of nature’s elegance.


But these guys are net pulling this lame but effective strategy that..am I missing something? Of course I am fighting to get stuff back that this awful information…
Hmmm.

Anyway non conflict is not viable. I am hoping the state and academia and medicine are reasonable. The private sector is …I have a marijuana company and vegan sweets i am proud to support. And i buy other things but that is still just goods and services.

These guys cannot start or run lame shadow societies. Or rather they can but they have to contend with other humans who exist like I do!

I exist! I swear I am not bigfoot ans I don’t think 5G kills birds or that SkyNet is on the horizon
But i do think this is bullshit and creepy, inhumane, crass, exploitative bullshit at that. And liable parties have to deal with consequences. I am a complainant and I am quite sure no one objects to complainants as much as to the creeps running these shows.

All this surveillance and narcotics bs..It is just a nightmare. But well…it just looks stupid and absurd as always.

[hunter]

11. AI is a front for something far more insidious.

As it exists now, what they are calling "Artificial Intelligence" is a stupid party trick. That's going to become apparent soon, if not already. People's belief and reliance on this technology will inevitably lead to some great catastrophe -- hopefully financial, not deadly.

Unfortunately all these data centers that are being built will not be torn down and their components dispersed. Instead they will be turned into automated surveillance systems more intrusive than anything George Orwell might have imagined.

Big Brother will be watching you always. You won't be able to take a piss without the time and place being recorded in some database.

[jfz9580m]

16. This is true

But it is also lame and stupid. I feel that part is being left out wrt authoritarian societies and fascism. And it is a very incomplete picture without that.
My dad used to draw caricatures to relieve work frustrations…I suppose in a way I am in the ballpark…

This stuff is so stupid and it is such junk..It is frustrating to watch once you are caught up. That’s the thing…a human brain gets the picture faster than those dumb things..But then if that advantage is worked against..which must not happen again.
These guys are now in the spaces of the civilian population..I protest this !

You can’t reason or deal with the creeps or morons.with the conflict that isn’t about stupidity or creepiness, but yet again screwing us over…i don’t even they can always help it.
Things have become so bad they are willing to allow us to exist..but as if one has to vociferously defend one’s right to exist ans explain how you want this foul thing, ALL of which i wish was non existent as far as the tech goes, ..
I don’t want this shit.


I agree with you about it being a front for something insidious. That keeps changing too fast and is now easy to identify ..but it took 14.5 years. Btw Hunter..i think there is a small chance I am in that stupid Stanford’s 100 years of ai research subject pool, because I worked at that stupid tacky place while this nightmarish douchebag called Hennessy was president.
So I consider ai extra lame.

My mentor was cool and some stray mild looking randoms seemed okay, but as a straight talking, tough, middle of the road mom of 6 that I am not, I generally protest all those other idiots! It is a joke and kind of true..

Those people inflict their vogon poetry on things…No..not really…

[cstanleytech]

12. I suspect AI is going to help us far more.

If anything I think it'll be more of a supplant to help support us such as helping doctors catch problems sooner with a patient's treatment or helping power companies reroute energy and help us monitor things like sudden weather events sooner like tornadoes.
As for the unemployment all they need is to just adjust the retirement system to one where people can truly afford to retire rather than the current system of working them to death and that change alone would open up plenty of jobs.

[hunter]

20. AI models of the electric grid really ought to be air-gapped from the grid itself...

... same as actual control systems of the grid ought to be air-gapped from the internet wilderness.

Depending on AI to control the grid seems a recipe for major life-threatening multi-billion dollar catastrophes.

We're not being careful enough with AI in medicine either. Often the medical professionals who are hired to "supervise" AI are simply fall guys, someone who can be fired (or worse...) when something goes wrong.

Ideally, electric grids and health care systems ought to be designed to have some high level of inherent stability and resilience.

I think EVERYONE deserves a secure comfortable place to live, clean water, healthy food, and appropriate medical care, not just retirees and the employed.

[questionseverything]

14. Ai burns through fresh water 💦, a fundamental element of life

So ty Bernie for trying to slow down the madness

[tavernier]

19. In my limited experience, AI is like a friend looking for your approval

who will model the answer to your question based on the way you ask it, seeking to give you an answer that will fit the result you are seeking. This is nice and cozy between third grader BFFs, but not helpful, and potentially dangerous, when seeking solid advice.