By Marc Steiner
Copyright therealnews
Republish this articleThis work is licensed under a Creative Commons Attribution-NoDerivatives 4.0 International License.
We encourage republication of our original content. Please copy the HTML code in the textbox below, preserving the attribution and link to the article’s original location, and only make minor cosmetic edits to the content on your site.Trump and Silicon Valley’s plan to rule the world with AI weapons
by Marc Steiner, The Real News Network September 9, 2025
Trump and Silicon Valley’s plan to rule the world with AI weapons
by Marc Steiner, The Real News Network September 9, 2025
The unnerving alliance between Big Tech and the second Trump administration goes much deeper than many in the public realize. What factors are motivating President Trump and Silicon Valley’s monumental investments in “artificial intelligence” infrastructure and the expanded integration of “AI” into military and governmental operating systems? What is their endgame? What will their vision of an AI future mean for the rest of us? In his new investigative series Military AI Watch, produced in partnership with Project Censored, award-winning science reporter Peter Byrne explores “how Silicon Valley, corporate media, the Department of Defense, the banking industry, and scientific institutions all intersect in the effort to militarize AI.” In this episode of The Marc Steiner Show, Marc speaks with Byrne and TRNN Editor-in-Chief Maximillian Alvarez about “AI” today and the techno-dystopian future on the horizon.
Peter Byrne is an award-winning investigative reporter and science writer based in Northern California. He specializes in reporting on physics, nuclear warfare, medical research, local and national politics, corporate and governmental corruption, and, now, the weaponization of artificial intelligence. Byrne’s work has been published in Scientific American, New Scientist, Quanta, Wired, Mother Jones, SF Weekly, CounterPunch, The Edge, Project Censored, and other publications. Byrne has published books and chapters on multiple universes and the history of quantum physics with Oxford University Press, Princeton University Press, and Springer.
Additional resources:
Project Censored, Military AI Watch
Peter Byrne, Project Censored, “One Ring to Rule Them All: Silicon Valley’ plan to conquer the world with AI weapons. Can we do anything about it?”
Maximillian Alvarez, The Real News Network, “‘Microsoft is an active partner in the genocide!’: Inside the tech worker revolt for Palestine”
Maximillian Alvarez, The Real News Network, “A bitcoin mine in Texas is “killing us slowly,” local residents say”
Maximillian Alvarez, The Real News Network, “AI data centers are draining water from this drought-stricken Mexican town”
Harry Davies & Yuval Abraham, The Guardian, “‘A million calls an hour’: Israel relying on Microsoft cloud for expansive surveillance of Palestinians”
Producer: Rosette Sewali
Studio Production: David Hebden
Audio Post-Production: Stephen Frank
The following is a rushed transcript and may contain errors. A proofread version will be made available as soon as possible.
Marc Steiner:
Welcome to the Marc Steiner Show here on The Real News. I’m Marc Steiner. It’s great to have you all with us. Now, we’ve all heard about artificial intelligence. Sometimes it’s just taken for granted as part of the modern landscape and our films and literature can seem ominous and dangerous, but in reality it can be both. But what happens when artificial intelligence is weaponized by the military, by our government, which occurs when right-wing tech giants control this industry in a time when the right controls our government. But I guess today’s Peter Byrne, he’s taken on this question. He’s an award-winning national and international investigative journalist and wrote an in-depth series for Project Censored, exploring every possible aspect. And in this conversation I’m joined in the studio by our editor-in-chief and colleague and dear friend Max Alvarez, making sure I don’t screw up. No, I’m just kidding. He’s here because we both are really intensely interested in this topic. And Peter, I want to welcome you. Thank you so much for joining us here today.
Peter Byrne:
Oh, thanks for having me. Glad to be here.
Marc Steiner:
So where do we even begin? I mean, reading this series you did. I figured we could do a five part conversation for this whole series, but let’s take a step back for a moment. When people hear that word, artificial intelligence, people have real more comprehension about the depth of it, what it means. So talk a bit about what that means, first of all, and what made you jump into this?
Peter Byrne:
Well, artificial intelligence, so called is probably one of the cleverest and inherently evil terms ever to emerge out of the military, industrial, academic complex as it were. It was coined in the 1950s at a conference at Dartmouth University where a bunch of cyber Ians who were early computer scientists got together to talk about the future of what we now call artificial intelligence. They called it computation and cybernetics. So anyway, they came up with this term artificial intelligence, which has stuck ever since. And in my newest story called Into the Uncanny Valley human AI war machines. I started by quoting David Hume, the philosopher from the 1757. He says There is a universal tendency among mankind to think of all things as being like themselves and to transfer to every object, the qualities they are familiarly acquainted with. We find human faces in the moon, armies in the clouds, and if not corrected by experience and reflection, our natural tendency leads us to ascribe malice or goodwill to everything that hurts or pleases us.
And to put things in context, I follow that with a quote from a military AI scholar. There is such a thing, the subject of military AI is not well covered at all in the mainstream media or even the alternative press. It is a function basically of trade journals which are pumping it up and our military AI watch at Project Censored, which is why I’m working there because it’s actually quite censored in the mainstream. Anyway, this military AI scholar referenced hues remarkable insight, which I think is applicable to using the word like intelligence for computation. And he said, AI technology is already being infused into military machines designed to look and act like humans. Anthropomorphic terms like ethical, intelligent, and responsible in the context of machines can lead to false tropes, implying that inanimate AI agents are more capable of moral reasoning, compassion, empathy, and mercy, and thus might act more ethically in humanly in warfare than humans.
And that’s the claim that is being made by Silicon Valley, which is just battening off this vast cornucopia of funding that’s being generated for a technology that is actually failing in many ways, which is that ais can replace, usis will be kinder and gentler than us, and by the way, we can also program them to kill our enemies at speeds faster than we can contemplate ourselves. There’s a whole school of thought that super intelligent ais are going to take over and exterminate human beings. That’s a conjecture. I don’t really go into that in my reporting. I’m consuming more with the here and now. And the dangers of the here and now are plenty because the present day is horrible enough in terms of one, how artificially intelligent so-called weapons systems are being used right now in Gaza as Israeli Palestinian Journal of 9 7 9 has exposed and has gotten pretty good coverage over the last few months.
The Israeli armies are using Microsoft and Oracle and probably open AI and other products to target people and to then kill them with so-called smart bombs. But one of the problems of course is, well, there’s two problems. One that they’re doing it at all using the databases that they’ve been putting together for years because they’ve been surveilling everybody in Gaza for decades. But two, there’s still obvious false positives. I mean, every time they want to kill some alleged Hamas leader, they end up blowing up a whole block full of children and people going about their daily business trying to find food among a starving genocide situation. So the Americans are not only complicit in the genocide itself, but in the technology of it because it’s Silicon Valley that is making it possible. And that’s only one application. We see military and intelligence applications of it everywhere from license plate monitoring to ice, try to figure out where to go to apprehend families of so-called undocumented people.
And of course the media gets all excited. The New York Times is just like the biggest blowjob of all on ai. I mean those guys are Silicon Valley’s just always astonishing them by the miraculous stuff that they can do. So my series goes into the tech of it because I am a science writer, but I’m also an investigative reporter. So it goes into the financing of it a lot. The first story, one ring to rule them all goes, starts out a long piece. It’s like 7,000 words. It goes into Palantir and Andre, these companies founded by Peter Thiel who’s a,
Marc Steiner:
Lemme stop you for one second. Want to back up a minute what you just said here. I think it’s really important for people to understand as we dive into what this technology is doing and is capable of doing, is the names you’re throwing out, who they are, what they are, what is Planete, who are these companies and the relationship that Teal and others have with this, the cross a spectrum of Democrat and Republican. And I think that because we’re on the edge of something really can redefine things in our society. So let’s step back for a minute and really define some of these terms. Who are these companies and describe them?
Peter Byrne:
Back in 2003, this venture capitalist named Peter Thiel, T-H-I-E-L, who had made a few million dollars in PayPal along with Elon Musk and Reid Hoffman and a bunch of other people who became oligarchs AI oligarchs. He founded this company called Palantir with investment from the CIA’s investment arm, which is called in Q Tel. A lot of people don’t know that the CIA actually has been pumping hundreds of millions of dollars into Silicon Valley AI weapons programs for a long time
Marc Steiner:
Or that the CIA actually has those investments that they have that kind of money. People don’t realize that either.
Peter Byrne:
No, I mean they also don’t realize that the Army has the Defense Innovation Unit, which is doing the same thing for the Army. With the advent of the Trump regime, the second day after his inauguration, he gathered all these oligarchs, Sam Altman and Larry Ellison and the guy from SoftBank into the White House, and also we’re going to put 500 billion into data centers. So this huge bubble of AI has been created. Now how did we get there? So let’s go back. Like you suggested 2003, Palantir has created, what it is is a very good software application for finding patterns in large databases. One of the first customers was a national security agency, was scoops up vast amounts of data every day and has it’s impossible to monitor it in real time. So Pound Share’s product is able to sit on top of vast amounts of data and locate where is the people wearing kafis today or where were they two years ago?
It doesn’t do much more than that, but that’s a lot from the intelligence agencies. And in order to weaponize this capability further, teal started another company called Andre, which manufactures the hardware that can use the Palantir type location software to go and find people and blow them up. Both Palantir and Andre were private companies. They were named after Tropes and Tolkiens Lord of the Rings. And there were EDRi still private. Palantir was private for 20 years. And what his business model was, which I expose in the first part of my series, is it was always running at a loss. It was always spending more money than it had, which had a real benefit for it because then it could tell the IRS that it wasn’t making any money and it didn’t have to pay any taxes. These guys hate to pay taxes more than anything in the world.
They contemplate having floating islands in the ocean so they don’t have to pay taxes. They get behind a guy like Trump who doesn’t want billionaires to pay taxes. So they would run out of loss and they were going to go back to their investors and get more money. And then when they went public a few years ago, we were able to using SEC filings, look into some of the machinations that they were doing financially. The short story of the long story is that now with the advent of the Trump regime, it’s finally making money and it hadn’t been making money before. Let
Marc Steiner:
Stop you for a second. So what do you mean? Why are they making money now? And did one before I shortly. What does that mean?
Peter Byrne:
Because they’re stuffing Palantir and Andre with contracts almost every day. They get a new contract, they’ve got billions and billions of contracts. Andre, which I point out in my first story in the series, has been getting all these contracts from the Department of Defense build as serving small business administration contracts. It’s not a small business. The reason they do that is because when you get a small business contract, you don’t have to go through all the procurement documentation and auditing that Lockheed would have to go through. Not that matters that much because the Pentagon can’t be audited, never has been audited. So at any rate, this is now on steroids. I mean, the reason that Zuckerberg and Tim Cook from Apple and the rest of ’em have been flocking to bend the need to Trump is because the government is throwing hundreds of billions of dollars at them, including in this so-called Golden Dome anti-ballistic missile program, which is scientifically and technologically a shival. If it cannot work, it cannot work. But they don’t care because what people need to realize not only about Silicon Valley, but about what’s going on with the Counter Revolution led by the Trump regime is these people have no ideology. They have no actual goal except enriching themselves in concentrating capital more and more and more, and that’s what they’re about.
Maximillian Alvarez:
I want to kind of tug at that thread a little more, right, because we’ve got two concrete horrifying examples that have already come up in the conversation so far, and I want to take advantage of Peter’s expertise and all the deep research he’s done. But for our listeners right now, I wanted to bring in both how AI and big tech are deeply embedded in the ongoing genocide of Palestinians and Gaza and ethnic cleansing of Palestinians in the West Bank right here, right now. And I also want to look at how the new Trump administration is leaning into big tech, actually not leaning in, like going as Peter said, full bore
Into partnerships with big tech. I want us to sort of unpack what they’re doing, how they’re doing it and why, what the end game is for them. But to kind of quickly go back to Israel and Palestine, this is connected to a story that we just reported on. I flew out to Seattle, Washington and I was one of the only journalists on the ground at Microsoft’s headquarters in Redmond, Washington when current and former tech workers at Microsoft community supporters, people with the no Azure for Apartheid coalition have been waging really risky, dangerous, brave, and crucial acts of resistance in the heart of Microsoft in direct opposition to Microsoft’s complicity with Israel’s genocide of Palestinians. And this was really exposed in an explosive investigation by The Guardian and the Israeli Palestinian publication plus 9 7 2 magazine and the Hebrew Language Outlet local call, and I’m going to quote from their August report here, which we’ll link to in the show notes, but the authors write One afternoon in late 2021, Microsoft’s chief executive Satya Nadella met with the commander of Israel’s military surveillance agency unit 8,200 on the Spy Chief’s agenda, moving vast amounts of top secret intelligence material into the US company’s cloud meeting at Microsoft’s headquarters near Seattle, a former chicken farm termed high-tech campus.
The Spy Master Yossi Sadel won nala’s support for a plan that would grant unit 8,200 access to a customized and segregated area within Microsoft’s Azure cloud platform. Armed with Azure’s near limitless storage capacity, unit 8,200 began building a powerful new mass surveillance tool, a sweeping, an intrusive system that collects and stores recordings of millions of mobile phone calls made each day by Palestinians and Gaza and the West Bank. According to three unit, 8,200 sources, the cloud-based storage platform has facilitated the preparation of deadly airstrikes and has shaped military operations in Gaza and the West Bank files suggest that by July of this year, 11,500 terabytes of Israeli military data equivalent to approximately 200 million hours of audio was held in Microsoft’s Azure servers in the Netherlands while a smaller proportion was stored in Ireland. Now again, people can go to the real news to look at our coverage of the Microsoft workers current and former who are putting their jobs safety on the line to oppose this however they can.
Microsoft is the most complicit digital arms manufacturer in Israel’s genocide of Palestine, Reen Ott, a current Microsoft worker and no Azure for apartheid organizer told me. And then Julius Sean, a software engineer at Microsoft also told me it really, really disgusted me to know that my work truly is in some part condemning children to death and I could no longer live with that. So with all that upfront, I wanted to turn things back to Peter and asked Peter if we could use these two again horrifying examples of big text complicity with Israel’s war on Palestine and the marriage between the current Trump administration and Silicon Valley or Surveillance Valley as you refer to it in your report. Let’s use these as kind of two critical examples to unpack some of the really critical takeaway points that come from your reporting here.
Peter Byrne:
Well, Gaza is the ground zero right now. It’s a laboratory for testing these weapons and it’s not just Microsoft. Amazon web services is implicated just as much as Oracle and all of the biggies. When you were reading from plus nine, seven two’s incredible reporting, you referenced the cloud. And when people think of the cloud, this is another marketing thing like artificial intelligence, right? It’s the smart as us. The cloud is up there, it’s like floating up there. No, it’s data centers that are being built just like at a pace that is astonishing. We talked about Stargate, where Trump had these guys in the Oval office and said they would put 500 billion into these data centers. They are death on the environment. They suck up vast amounts of water to keep the chips cool, and they use vast amounts of electricity, which is why Amazon and the rest of ’em are trying to build them in deserts because building these data centers is really capital intensive and places like OpenAI, which is basically just a division of Microsoft these days, they don’t have the money to build the data center.
So they have to go into debt, they have to borrow money from M SoftBank, which is this semi fraudulent organization based in Japan, which has been in a lot of trouble, but which is at the center of everything because it’s been sucking up to the Trump and his crypto moves. So we have these data centers that are being built and one of my articles about the data centers I talk about, yeah, those are going to be actual ground zero in a nuclear war or even in a conventional war because the wars that use autonomous weapons and AI control systems are actually all taking place inside the data centers inside the chips. So you don’t need to just attack a plane or a ship, you could just go blow up the data center and those things will be mean entering around like lost spot. So you’ve got the data centers and then you’ve got, as one of my other articles called AI battlespace, weaponizing the 5G Internet of Things.
You’ve got what the military calls the military internet of things, and this goes back to Amazon and Alexa and all of these devices like your refrigerator now, which has a chip in it that utilize the 5G electromagnetic infrastructure that is now being implemented worldwide. I quote from Rand reports and military documents showing how the military is now planning to use the 5G devices in households in Europe to guide its AI weapons systems in the event of a war with Russia. They’re actually, depending on all these little devices that we think are making our lives easier to make military operations happen in places where there’s so much jamming going on because that’s a whole nother story. It’s easy to jam the electromagnetic spectrum and to mess up the AI’s capability of communicating each other, but if they can just switch from communicating through satellite constellations to communicating through the Alexas in people’s houses or their refrigerators and smart toasters, which they actually can do, you don’t have to give them permission to do it, they can override it, then that’s how they plan to fight these future wars.
Now I want to make two points. One is the internet of things and all the sensors that they have and the satellites, I mean we’re talking about millions of devices all over the world that are supposedly sending information into some kind of centralized point in the military industrial apparatus from which they can control everything, and that is their plan. They have what is called a joint all awareness domain command and control system, which they envision as a kind of pane of glass where they can see the entire global battle space as they call it, and they can wage wars big and small, including nuclear wars at light speed, all basically operated by eyes with little or no human input. This is their wet dream. This is their vision because it’s costing a trillion dollars to implement. Now, the good news from the point of view of the actual humans is that it probably won’t happen for several reasons.
One, the military typically the Army has its contractors. The Air Force has, its the navies has its, and it’s the same thing going on with Silicon Valley. They each have their own favorite. I think it’s the Army likes Amazon and the Air Force likes Oracle, I think, but they have, they’re signing their own definite contracts with each of these big Silicon Valley oligarchical companies to make this great big joint all awareness domain and control system. But it’ll be proprietary only to that army or that Navy or that air force. And they don’t interconnect, they don’t talk to each other. So there’s no way that the president or the civilian who’s supposed to be giving all the commands, it is going to be able to look at a central point and go, oh, we’ve got a little battle going on in Bosnia. Let’s hope it doesn’t.
Our resources need to go there, but we have resources in Pakistan and we’re fighting a war with China in Taiwan. There’s no way that there’s any kind of centralized decision making can happen because these things can’t talk to each other. But even more fundamentally, they’re all based on large language models, the so-called chats, and the chats hallucinate all the time. This is an inescapable feature. They confabulate because they’re predictive mechanisms, they have no sense of truth. All they can do is to look at their training data, which is usually about selected by the military forces. They select the training data that they want them to train on. They don’t talk to about peace or diplomacy or arms limitations or anything like that. And I go into a lot of that in my articles. So you’ve got these chats that are advising military commanders in theory and in many cases will actually be issuing tactical orders that can’t tell truth from falsehood.
Now, a lot of people in the Pentagon are horrified by this. You’re weird Christian Dash crypto fascist leader, Pete Heg and Trump’s AI czar who was also one of the PayPal mafia guys. Their concern, as I said earlier on in our talk today is money. They don’t care if his shit works or not, but in order to get the money out of Congress and the American people, they have to go, oh, China’s going to beat us in the AI race and then they’re going to dominate us and we’ll all be speaking Chinese, blah, blah, blah, drumbeat going on day after day. We have to get ready for a war with China. Well, this is no joke because every time they want to go to war, remember Iraq and Afghanistan, I mean it just goes back on and on and on. They start the drumbeat. Our way of life is being threatened by this is an existential issue. So they’re throwing all this money into a technology that the science itself shows is incapable of accomplishing what they say it can’t accomplish. It’s one of the largest fraudulent speculative bubbles in human history and it can only harm us in ways that we can’t even yet imagine.
Marc Steiner:
So just to push that a little bit further, I mean two of the things I gleaned out of the work you’ve done and thinking about this over the last few days is that if you take the two examples of ice and how the technology’s being used by ice at this moment to go after immigrants coming to this country, and if you look at what is going on in Israel, Palestine and Gaza, that these are both testing grounds because what they’re doing at this moment in those two fronts that I’m talking about, they’re using this AI to kind of further their goals. It’s a test they’re working on implementing this and building on it to create a structure that doesn’t exist now, but doesn’t mean it could not exist. And when you take that with the corporate world that is helping to fund and push this and the right wing push across the world, especially in our own country, what I glean from this is the Amish potential inside of our future with all this.
Peter Byrne:
I’m complete a chord with you on that in the sense that yes, it is a test. Palestine is famously a laboratory, but right now in the United States, they’re testing Palantir and ail are both totally embedded with ice. They have billions of dollars in contracts right now to do what it is. When I say that it doesn’t work technologically, it doesn’t work as they bill it. For example, they say, well, we’ll be able to tell the difference between a peasant holding a rake and a rifle from our satellite based ai. No, that’s not true. I mean, we know that there are limitations on that. Now the idea that large language models, which is just one type of artificial intelligence, which we should be clear about is the latest iteration. There are other types of artificial intelligence. I want you to actually much better at doing this type of computational analysis than large language models, which are crippled by their hallucinatory problem.
They’ve also run out of data to train them on. They used up a whole internet. They sucked up Reddit and Wikipedia and the common crawl, just a lot of internet garbage. They didn’t get into university data banks or corporate data banks or even military data banks. I mean, those are all things that can be fine tuned into them, but the foundational models are all trained on just internet slop. So that’s one of the reasons that they hallucinate, but they run out of data. They need massive amounts of data to make their probabilistic neural nets work, and there’s a lot of papers that are coming out, they call it at the ceiling, they hit the ceiling and they’re not going to get any better at what they do.
Marc Steiner:
What do you mean? What do you mean they’re not going to get any better? Why do you say that?
Peter Byrne:
Their problem is it’s garbage in, garbage out. It’s the training data, but like I mentioned before in Gaza, if it was doing what they said it would do, it would be able to say, okay, commander so-and-so of hummus is in this apartment, and we can go there and either arrest him or pinpoint a small explosive and blow up his apartment. But no, they blow up the whole block, right? Because they don’t actually know if he’s there or not. The claims that are being made by the chats are probabilistic. They’re not factual. They’re saying there’s a chance that so-and-so is there or whatever.
Maximillian Alvarez:
I want hover on that just really quick. I was
Peter Byrne:
Maximillian Alvarez:
Go ahead. Yeah, I mean further unpack this for our listeners, right? Because I think help make Peter’s point about the limits of what this technology is going to be able to do. It’s worth, like I said, hovering as we are right now on what it is currently being used for and the results of that. So between Gaza and the West Bank, let’s not forget that a lot of these surveillance efforts that have been turned on Gaza started in the West Bank and a lot of the data that has been just stored from surveilling entire populations of Palestinians. To Peter’s point, the general who kickstarted all of this in Israel got fired because that shit didn’t predict or stop October 7th, and that was the main thing that got him ousted. But what it is being used for as Peter was just saying, is to compile hundreds of millions of hours of phone calls, texts from, I’ve seen pictures of these data like collectors and surveillance scrapers that are set up in settlements in the West Bank that are just there to surveil and collect all of the communications data of Palestinians in those areas.
What we can sort of extrude from that is that right now with the different kinds of technologies that we are currently calling ai, we have tracking and surveillance capabilities. That’s one of the key use cases that is being pushed from Microsoft to the ID to the Trump administration. So you can scrape vast amounts of social media accounts for any language that is deemed to be supporting terrorist organizations or what have you, right? You have massive data processing capacities and ostensible pattern location capabilities. To Peter’s point, you feed massive amounts of data into these large language learning models for the purpose of being able to identify patterns that like the human eye and brain couldn’t, but that is distinct from the predictive and decision-making capacities that they are trying to sell from the IDF saying, we have all this data, we haveis going through it to locate patterns and we’re going to use AI to basically tell us who to target to kill.
Peter Byrne:
Let me interject with this, and I’m in agreement with what you’re saying In Gaza, there’s no counterforce that can jam the ability of the IDF to utilize the electromagnetic spectrum for all of the military intelligence uses that it does, but the Achilles heel of AI driven warfare and autonomous weapons is the electromagnetic index spectrum. The case in point being in the Ukraine where you do have powers that can jam the electromagnetic index spectrum and the Ukraine, there’s like 10,000 drones a day going in each direction and only a handful of them get through because both sides are jamming. The other side’s capabilities to guide their weapons. They’re frying the insides of the weapons with microwave direct devices, and so both sides have resorted to sending out drones that extrude miles of very thin fiber optic cable that can’t be jammed. It goes to where a soldier operator has a device to guide the drone into its target.
Most of the drones that are actually hitting targets these days are guided by fiber optic cables, and then they rely on the phenomenon of large numbers. If you send out a thousand drones, maybe six of ’em hit. So then you can claim that you’re doing well that day militarily, but you just sent out 900 whatever drones they didn’t hit, right? The thing is that Gaza, which is the laboratory, nobody’s jamming the Israelis, but what they’re doing there won’t work in other places where you have powers that can actually do the jamming and the GAZA’S can’t trick. The large language models that the Israelis are training are using, which are trained on data by poisoning Rand to the Department of Defense, and everybody else that knows about this are freaking out about how easy it is to hack the training data that large language models are using.
Just think about it, they scoop up trillions and trillions of bytes from the internet. So you have your little website, whether you’re North Korea or with anonymous or something, and you put some data poisoning code that gets scooped up. Then just for an example, if you changed two pixels on a stop sign, a large language model might think it’s an elephant. That’s a famous example. So the people with the technological capacity to poison and disrupt and hack these systems, which they are doing, render them pretty ineffective. I mean, I actually have a subscription to Jane’s, which is the trade magazine for all the weapons freaks
And time after time when they’re writing about AI battle equipment. It’s the jamming that’s the problem. It’s the poisoning. That’s the problem. I go into this a lot and my articles, so when I say that it is got a ceiling on training and it’s also got a ceiling on operational capacities given technologies that we have so far, pretty much I would predict that in a major war between the United States and China, which we used autonomous weapons, the unintended consequences would be just amazing because nothing will work the way that they think it can work, and they have no capacity to view the overall battle space. As I mentioned, they don’t have this joint all awareness domain capacity, and it’s not just me saying this, this is a government accountability office. I linked to reports by federal auditors saying exactly what I’m telling you.
Maximillian Alvarez:
I think that really helps clarify for me, Peter’s point about the entire promise of Silicon Valley of what this tech is going to mean for warfare and national defense in the 21st century is bunk because so much of the learning and data processing that goes into this can be jammed from the start and thus the processing and prediction are going to be flawed. So you’re not going to have these pinpointed targets of military leaders who are killed by an autonomously controlled weapon. That’s what they’re pitching. That’s not what they’re going to get, especially with any major nation that is going to war in the 21st century like China or the us. However, I want to kind of circle back to what it will and is being used for, especially when it’s turned on its own citizens, right? We’re talking about, I mean, this was what I think we’re going to keep learning for years and months, what Elon Musk’s team at Doge, what public data was fed into these language models without our consent, but that is now being used to track surveil and target undocumented or political dissidents who have attended protests. Everyone’s trying to implement this stuff to the point of denying medical claims using ai. So I wanted to sort of circle back there, Peter and ask, with all the ceiling that you talk about in terms of its military and intelligence capacities, let’s talk about how it’s already being used even with those problems you talked about to surveil and punish working people.
Peter Byrne:
I’m not sure that the AI products, the Palantir is put in place. It’s helping ice all that much because they seem to be cruising the streets, looking for people that have brown skin and maybe are walking out of a factory and then attacking them. And since people are undocumented and they’re not in that many databases necessarily, I’m not sure that the plan to be able to just identify millions of unwanted immigrants via data is working out that well for them. Mostly they seem to be trolling to get it. Now, when I say that this stuff doesn’t work as it’s flawed to be working, it doesn’t mean that it doesn’t have tremendous harm and that it doesn’t work in some ways. For example, I got a whole story a couple of weeks ago here on how AI has been inserted into the nuclear command and control system.
It’s actually, this is happening. The chats are being inserted into the so-called NC three, which is the nuclear command and control system, and war games and experiments and everything else show that because of their training, the chats almost always launch nuclear weapons in any type of conflict because that’s the bias that they’re trained on because they’re trained by the military and the military is biased to actually go to war, and the corporations are biased to go to war too. How you make money. I mean, Lockheed figured out long ago, we need to have wars in order to keep our stockholders happy, and Peter Thiel has figured out the same thing. So putting it into the nuclear command and control system is like an act of insanity, but how much more insane is it than we actually have a nuclear command and control system to start with, and that we have these weapons that we’re not even trying to control anymore. We’ve abandoned the start. We’ve abandoned all anti-ballistic missile treaties. We’re headed towards what the United States is now advocating, which is the ability to fight limited nuclear wars, which is what Henry Kissinger was advocating back in the 1950s and was famously shot down, but now it’s back because the tech, we have smaller bombs, so they think they can fight these wars and not end up with nuclear winner and devastation and stuff like that.
I think that everybody’s got PTSD and basic mental instabilities in the world now. I mean, not only because we’ve evolved technologically passed our homeowner capacities to process information, but because we’re all staring into these fricking phones being told what to do every day and we’re being fed just this steady diet of spectacle. And so the horror here is that a lot of people think one, that it’s okay, what’s going on in Gaza, and two, that we should go to a war with China and three that Trump’s doing a good job or that Biden was doing a good job. He was just as equally evil, and so how do we reorient ourselves culturally and politically in this vast morass of misinformation and disinformation that is targeting us? And that’s where ais are really good. Is it targeting you individually with misinformation which somebody is paying them to do?
However, the good news is there’s a bunch of reports came out in the past couple of weeks showing that businesses that bought in and drank the Kool-Aid and fired a bunch of people and replaced them with AI are now hiring people back because things aren’t working very well, because like I said before, they hallucinate. You don’t want your customer service AI to go, Hey, fuck you buddy. I think your wife’s a bitch. I mean, they can say stuff like that. They really can. And so the companies are running away from it right now, which is super interesting. I think we’re looking at another AI winner. The bubble’s going to collapse this year. I would predict, I would bet money on it. And that doesn’t mean that the next technological iteration building on the neural net configuration that we have now won’t be more capable of doing this or that. We don’t know. It’s all conjecture, right? But symbolic ai, which was the kind that came before the large language models, is actually much more accurate and rules-based as it were, than the predictive models, which basically have no rules.
Marc Steiner:
I just kind of push back a little. Everything I’ve read in all the pieces you wrote and other things I’ve been looking at, we’re in kind of very beginning stages of this technology and all of it may not work at this moment. That’s exactly what they’re doing in Gaza. That’s exactly what they’re doing in ice. That’s exactly what they’re doing in other situations is testing out these models. How does it work? Where does it fail? How do we make it stronger? And when you’re doing this in encountered with very right-wing governments like our own in this country and with the technology companies that are allied with that politically, this is to me just the beginning. We don’t know where they’re, this is going to go. We don’t know where we’re going to take it, where they’re going to take it. And everything I’m watching at this moment are like dry runs. Is this going to work? How do we make it work and how do we make it better so it does work.
Maximillian Alvarez:
And to even add one more horrifying detail to that, I think what we’re all also acknowledging is the fascistic reality that we will base life or death situations on the hallucinations of an unaccountable technology in the hands of the rich and powerful. So when AI is being used to track surveil process data from Palestinians and target Hamas fighters, and it results in women and children being blown up, that’s as horrifying as the technology because the technology doesn’t work as it is, as if the technology did, right? I mean,
Peter Byrne:
Let me interject here. I’m not saying that it doesn’t work. I’m saying that it doesn’t do what is being sold as doing, which is to be discriminatory and obey the international humanitarian laws of war and things like that. I mean, we have to recognize that ai, artificial intelligence has actually been embedded in the military systems of the United States and the other Western industrial countries for like 70 years. I mean ever since the bomb sites in World War ii in the B seventeens, those were artificial intelligence, the origins of it. So we’ve been computerizing our military apparatus forever. This is just the newest iteration of it. And yes, it is being tested. The testing process itself is horrible in every ways. If you want to glimpse the future that Silicon Valley is aiming towards, look at what’s going on with the Uyghurs in China, their entire 2 million strong society, it seems to can be completely monitored by devices set up by the Chinese government that in conjunction with Apple and Microsoft and Oracle and the rest of Silicon Valley, they would like total control.
Just like Leonard Cohen said, they want absolute control over every living soul. But that doesn’t mean they’re going to get it because people need to say, I’m not going to go anymore with this. I’m going to put down my phone or not going to use the chat. I mean, using the chat is an environmental crime every time you do it, but not only that, they’re using the information that you feed the chat to train more chats. So it does bad things and it is limited, and it will just keep doing bad things as it struggles to keep making profits because that’s the main goal of the war machine. This guy, Antoine Buske wrote this. B-O-U-S-Q-E-T wrote this great book called The Scientific Way of Warfare some years ago, which goes into starting with the Greeks, how technology and science have implicated warfare and developments of warfare up into the present day. So this isn’t new. Using ai so-called in warfare is not new. It’s just getting a lot more attention right now, and that may be a good thing. Maybe people will get sick of it and turn against it.
Marc Steiner:
So I think that one of the things after reading all the pieces you wrote and want to read more and get into this even deeper, is what the response is. What should our response be? How do you control it? How do you control them? How do you go up against the powers of techno business across this country and across the globe? It is a battleground for the future of democracy. It’s a battleground for what happens to our society because what they can’t do now, they will figure out a way to do it. And if I’m as old as I am, I may not be here, but my kids will be here, my grandkids will be here, my great grandkids will be here. And so I think it’s something that people really need to understand what’s happening to us, what they’re doing, what we face, and then how do we respond to it.
Maximillian Alvarez:
Well, and I want to just hop in with three little points there, and then I want to give Peter the last word, please on that really important question. But just taking from our own reporting here at the Real News, we’ve already seen that every step of the way there is resistance, and there are critical sites of resistance where working people are the ones fighting back either because their conscience is telling them they have to or because their circumstances leave them no other choice but to fight. So I interviewed folks in Texas who are living across the street from these Bitcoin mining data centers that just popped up in their neighborhoods, and now the sound pollution from those is giving people heart problems and they’re bleeding from their ears. And so you have these coalitions of apolitical people, liberal people, deep red, conservative people who are all being kind of pushed together because these data centers are, as Peter said, just so there’re such big resource hogs, they’re taking all of the water.
They are polluters sound wise and otherwise, and people don’t want to live near them and they don’t want to give all these resources away to these big tech companies that aren’t even providing jobs for those communities or any benefits that people can see. Seeing grassroots resistance around the data centers from Memphis to Texas to Mexico, that’s really critical. I mentioned the Microsoft workers and tech workers in general who have a vital role to play. They are the rank and file workers on board the Death Star in a lot of ways. And they know that the workers I talked to know that, and they are speaking up. They are disrupting business as usual. They’re using their knowledge to convince people of what we’re trying to kind of unpack here in this conversation. But the Microsoft are explicitly calling for a worker Intifada. They’re saying We need tech workers across the sector, not just in the US, to mobilize resistance against what the executives and shareholders are pushing these companies to do, including surveil and Target and kill Palestinians and our own people here in the United States.
So you have those two critical sites of resistance. And the last one I just throw in is that other workers who don’t work in tech also have a critical role to play. And we saw that coming to a head during the Hollywood strikes last year, particularly the writers strike one of the major, I mean, but this was also an issue for SAG AFT and the screen actors because they were fighting the company executives who are trying to push for AI to replace those workers. They want to create movies that are written by ai, maybe checked by one human. They want to replace background actors with AI generated ones. They want to use dead actors likenesses to recreate new performances. So this is coming, it’s seeping into all areas of life, which means that yes, it’s horrifying, but at the same time it creates more stakeholders in the fight against it. And I think that we’re seeing that from residents living near the data centers to workers in tech to workers whose jobs and livelihoods are being threatened by tech. And with that, I’m going to shut up and let our guest, Peter have the last word.
Peter Byrne:
Yeah, I’m in unity with what you’re saying. I think as people are more and more recognize that they’re locally impacted by the AI infrastructure, whether it’s data centers or losing jobs or getting bad medical information from their chat, whatever, that people will look for answers now. Yeah, general strike would be a good idea. I mean, if people stopped using Amazon for a week, that would change everything, which is why Amazon is actually, its profit center is now a military contracting. It’s not the Amazon store, but if people stopped using it, that would probably bring it to its knees. What we have to do, I think in this country is to look beyond the binary political thinking that we’ve been brainwashed to think it’s got to be either Democrats or Republicans. The Democrats are just as implicated in all of this as the Republicans. I mean, the Biden administration was going in the same route that the Trump is on ai.
There’s really no difference in what happened in the transition and the Tweedle D and Tweedle dumb two sides of the same coin kind of politics that we’ve been imprisoning ourselves with here for decades, if not centuries, doesn’t have a solution. There’s no way the Democrats are going to stop Trump from running for a third term or whatever he wants to do. So it’s going to take a mass movement and people just have to stop looking at their phones and stop thinking that they can communicate everything on signal and be immune from being attacked themselves and to actually start talking to one another face to face-to-face. I mean, look at us. I’d love to be in a room with you guys, but we’re here like a thousand miles, 3000 miles apart. And no doubt, the NSA has scooped up every word we said in some large language model is making an analysis of it.
Fortunately, it would be largely untrue. So I don’t know what’s going to happen next. I mean, mark, you and I are basically the same generation. We came up through the 1960s and all that, and people were mobilized by the depredations of the Vietnam War and by our encouragement of the Black Panther Party for self-defense and things like that, what do we have now? Black Lives Matter turned into some fricking nonprofit, self-aggrandizing hell, in terms of how it took on its next step. It didn’t really need a next step. It was great at mobilizing people as a phenomena, but once it became institutionalized, it became easy to attack. And so most things, movements that become institutionalized, become corrupted one way or another. So I don’t know. I have a 23-year-old son who’s a college senior, and for some reason I don’t understand, he wants to be an investigative reporter. So I think we need more people out there doing that and communicating. Then let’s see what happens. Because we’re not in control. The fucking war machine is in control, but it can’t do it without our blood.
Marc Steiner:
Well, this is just the tip of the iceberg that we’ve done here. First of all, I want to thank Peter Burn. Just we’re going to link to the piece you put together. It’s really important to read. I think there’s a great deal more to glean out of it and to talk about and to expand this conversation, which I think is critical. And
Maximillian Alvarez:
Shout out to our colleagues at Project Censored for all the great work that they’re doing.
Marc Steiner:
Absolutely. And to my dear friend, colleague and editor-in-chief, we had the Real News, Max Alvarez for being here on the Mark Steiner show as well in this conversation. And we’re going to look at this more and in greater, deeper ways because I think what Peter Byrne has shown us in his series is the real danger that we face and what we have to do to stand up to it, to save the future for our children, grandchildren, and everybody else on this planet. So thank you all for being here on the Mark Steiner show. Peter, thank you for doing the work you’re doing. And folks, let us know what you think. Just write to me at market the real news.com. I’ll get right back to you, and we’re going to expand this conversation and get it out there. So thank you both, all of everybody. Thank you,
Peter Byrne:
Onward. Thank you.
This article first appeared on The Real News Network and is republished here under a Creative Commons Attribution-NoDerivatives 4.0 International License.
