The Matt Walsh Show - May 27, 2026


Ep. 1786 - Tech Billionaires Are Openly Announcing Their Plans. Are You Ready for What’s Coming？


Episode Stats


Length

46 minutes

Words per minute

169.67441

Word count

7,857

Sentence count

419

Harmful content

Misogyny

1

sentences flagged

Toxicity

7

sentences flagged

Hate speech

4

sentences flagged


Summary

Summaries generated with gmurro/bart-large-finetuned-filtered-spotify-podcast-summ .

Transcript

Transcript generated with Whisper (turbo).
Misogyny classifications generated with MilaNLProc/bert-base-uncased-ear-misogyny .
Toxicity classifications generated with s-nlp/roberta_toxicity_classifier .
Hate speech classifications generated with facebook/roberta-hate-speech-dynabench-r4-target .
00:00:00.000 Ted Kaczynski, otherwise known as the Unabomber, was a mathematician who murdered and maimed
00:00:05.560 several people with mail bombs during a domestic terror campaign lasting nearly 17 years.
00:00:11.460 And Kaczynski's goal was to wage war on what he called the industrial technological system
00:00:15.940 and to return humanity to a much more primitive state of wild nature, where everybody lived like
00:00:21.220 he did, off the grid, in a cabin, in the woods, without water or electricity. And to prevent
00:00:28.040 further bombings, the New York Times and the Washington Post published Kaczynski's manifesto
00:00:31.660 called Industrial Society and Its Future at the urging of the FBI, which believed, correctly it
00:00:36.820 turns out, that somebody would recognize his writing style and turn him in. Now, what you
00:00:42.160 may not know about Kaczynski's manifesto is that it discusses the rise of artificial intelligence,
00:00:47.280 which is pretty remarkable for a document written in the early 1990s. Now, obviously, back then,
00:00:52.780 AI was nothing like what it is today, when students cheated on their homework or their
00:00:58.760 tests back in the good old days, they had to do it the honorable way. You know, raid the teacher's
00:01:03.940 desk, copy from the smart kid, fill in the answers on the Scantron sloppily so that it can't tell if
00:01:10.480 you filled in the A bubble or the B bubble, you know, stuff like that. I never did any of those
00:01:14.720 things, of course, but I heard about people doing it and I was really disappointed in them. Anyway,
00:01:19.100 But even back then, Kaczynski saw the potential of AI to cause mass disruption.
00:01:25.220 And here's what he wrote, quote, due to improved techniques, the elite will have greater control over the masses.
00:01:31.520 And because human work will no longer be necessary, the masses will be superfluous, a useless burden on the system.
00:01:37.780 If the elite is ruthless, they may simply decide to exterminate the mass of humanity.
00:01:42.280 If they are humane, they may use propaganda or other psychological or biological techniques to reduce the birth rate until the mass of humanity becomes extinct, leaving the world to the elite.
00:01:52.080 But suppose now that the computer scientists do not succeed in developing artificial intelligence so that human work remains necessary.
00:01:58.560 Even so, machines will take care of more and more of the simpler tasks so that there will be an increasing surplus of human workers at the lower levels of ability.
00:02:07.600 On those who are employed, ever-increasing demands will be placed.
00:02:11.160 They will need more and more training, more and more ability, and will have to be even
00:02:14.680 ever more reliable, conforming, and docile because they will be more and more like cells
00:02:20.180 of a giant organism.
00:02:23.160 Now, what immediately shocked people about this manifesto, regardless of how they felt
00:02:27.260 about the arguments themselves, to say nothing of the method Kaczynski used to get them published,
00:02:32.020 was how well-written and coherent it was.
00:02:35.460 As you've probably noticed, most manifestos written by murderers today are rambling and unreadable and really unimpressive.
00:02:45.040 Compare the cliched Reddit-tier manifesto of the attempted assassin at the White House Correspondents' Dinner to this.
00:02:52.060 They just don't make manifestos like they used to.
00:02:54.740 And that's why industrial society in its future found an audience.
00:02:57.620 In fact, it's still taught in some universities today.
00:03:00.540 Kaczynski made his case that as technology improved, people would gradually surrender
00:03:07.500 more and more of their civil liberties in order to make their lives more convenient to the point
00:03:11.640 that a violent authoritarian crackdown wouldn't even be necessary. Because of the success of the
00:03:17.460 manifesto for several years after Kaczynski was captured, there were concerns that copycats might
00:03:22.480 begin a new wave of domestic terrorism, if not the full-on revolution that Kaczynski demanded.
00:03:27.660 And those attacks never materialized, at least not to any significant degree.
00:03:31.960 But there are new signs that another Ted Kaczynski-style person may be on the way.
00:03:39.960 Wired Magazine just obtained more than 1,000 pages of unpublished reports from the Department of Homeland Security, FBI, and fusion centers,
00:03:47.640 indicating that the government believes AI-related terrorism is now a major threat.
00:03:52.820 And one of these Pennsylvania fusion centers, which connect federal and local law enforcement, warned recently that, quote, adversarial actors, including state-sponsored entities, criminal groups, and extremists, such as homegrown violent extremists or environmental extremists, may target U.S. data centers, and that these actors could also exploit the strategic importance of data centers to the U.S. economy, using them for activities like cryptocurrency mining or leveraging third-party entities, such as front companies, to gain access to U.S. data and infrastructure.
00:04:21.720 Wired also found a report from the New York Intelligence and Counterterrorism Bureau, quote,
00:04:26.660 The chaotic atmosphere that may result from emergent AI technology in the next five years may fuel large-scale protests that devolve into civil unrest and anti-tech violent extremist activity, especially in large urban areas such as New York City.
00:04:40.160 The report reads, the term anti-tech violent extremists does not appear in any publicly available DHS or FBI domestic extremism reports or guides and represents a novel grouping of a wide range of ideologies under a single extremist category.
00:04:54.620 In the same Intelligence Bureau assessment, analysts also described a novel threat emerging in the wake of the arrest and trial of Aziz Lasota, an extreme rationalist who allegedly led a small cult-like group, three members of which have been charged with murder, tied to an obsessive ideology focused on the existential risk posed by AI.
00:05:14.800 Okay, now the usual caveats apply here. The government has been known to label anyone they don't like as an extremist and to hype threats that don't exist by leaking documents to sympathetic media outlets.
00:05:28.820 Most recently, we saw how the Biden administration went after alleged Catholic extremists, parents complaining about their school boards, MAGA Republicans, January 6th protesters, pro-life activists, white men in general.
00:05:41.560 In fact, a couple of years ago, we talked about how the ADL formally demanded that the Washington State Fusion Center target me personally as an extremist.
00:05:50.260 So we really shouldn't put any stock whatsoever in vague government warnings about extremists, particularly when they don't offer any evidence.
00:05:58.260 At the same time, if our big tech overlords wanted to inspire a violent uprising, it's getting hard to imagine what exactly they'd be doing differently.
00:06:09.920 Now, to be clear, as I've said repeatedly, like all sane people, like shouldn't need to be said, but it does.
00:06:15.900 I oppose murder.
00:06:17.820 This has become something of a controversial take, particularly after the UnitedHealthcare CEO was shot to death on his way to work.
00:06:24.080 But it shouldn't be controversial because, you know, even putting aside like basic morality and human decency for a second, which we can't put that aside.
00:06:31.340 It's the end of civilization if we accept the idea that murdering civilians is somehow an acceptable response to policy disagreements.
00:06:39.100 Even if, in your estimation, more people will suffer and die unless your particular policy is adopted, which for the left basically amounts to making everything free and putting conservatives in prison, if not the morgue.
00:06:49.980 Once you settle your debates with violence, you lose the moral right and certainly credibility to object when your political opponents take away your vote and your property and ultimately your life.
00:07:02.460 your life. What we need to recognize, though, is that some level of violent backlash, even
00:07:08.480 terrorism, becomes sadly predictable when people are pushed far enough. That doesn't mean those
00:07:15.840 terrorists are morally justified. Of course, they're not. But it does mean that people
00:07:20.780 empower leaders, oligarchs, executives, and so on, run a significant risk when they start
00:07:26.800 antagonizing millions of Americans for no good reason. It's the same reason I think it's a bad
00:07:32.880 idea when CEOs get massive golden parachutes after destroying a company. Boeing CEO got something
00:07:38.600 like $30 million after overseeing multiple major airline crashes, to just give one example. Now,
00:07:44.540 did the board have the right to award that money? Yes, they did. I can even understand why the board
00:07:49.700 might have made a deal like that. But when people see enough of these golden parachutes for
00:07:55.300 incompetent executives while they're struggling to send their kids to school and to buy groceries
00:07:59.240 and get gas in their car, it's hard to blame those people for drifting towards, say, President AOC
00:08:06.720 and Vice President Bernie Sanders. Now, voting for that deranged ticket would be metaphorically
00:08:12.520 its own form of terrorism. But people do insane things and destructive things and self-destructive
00:08:19.420 things when they're pushed too far. That's the point. If you provoke angry, desperate people
00:08:23.560 enough, even if what you're doing is completely legal, and even if in some cases it's rational,
00:08:28.880 then you are inviting consequences that are much more far-reaching than you probably realize,
00:08:34.120 and everybody will feel the ramifications of it. So to give just one example of what I'm talking
00:08:39.220 about, here's the reception that former Google CEO Eric Schmidt received earlier this month at
00:08:44.660 the University of Arizona's commencement ceremony. And he repeatedly sings the praises of AI, and
00:08:49.900 the graduates start booing he seems to almost taunt them it's very bizarre watch last december
00:08:59.420 time magazine selected its person of the year for 2025 and it was this time it was the architects
00:09:05.740 of artificial intelligence interesting it will touch every profession every classroom every
00:09:15.580 hospital, every laboratory, every person, and every relationship you have. I know
00:09:21.880 what many of you are feeling about that. I can hear you. There is a fear.
00:09:32.060 We do not know, we do not know the precise contours of what this
00:09:39.400 transformation will look like. Choose a diversity of perspectives, including, let
00:09:45.320 Let me add, and if you'd let me make this point, please, if you don't care about science,
00:09:57.640 that's okay, because AI is going to touch everything else as well.
00:10:01.280 Whatever path you choose, AI will become part of how work is done.
00:10:07.100 If you have a problem in the world you want to solve, you can now assemble a team of AI
00:10:13.160 to help you with the parts that you could never accomplish on your own.
00:10:17.440 Let me give you some advice.
00:10:19.760 First, find a way to say yes.
00:10:23.340 And thank you very much and good night.
00:10:28.020 Now, I've been shouted down by mobs many times in my career.
00:10:32.440 It happened pretty recently.
00:10:33.820 I was at the Supreme Court after our major victory
00:10:36.340 against the sterilization and castration of children.
00:10:39.660 And as a result, as a general matter,
00:10:41.180 I find it extremely difficult to sympathize with mobs that scream at people who are trying to talk. 1.00
00:10:47.600 They come across as ignorant and annoying and pathetic. But also in this case, it's not hard 0.99
00:10:55.080 to see where the students are coming from. They shouldn't be jeering the commencement speaker, but
00:10:58.800 you can understand where they're coming from. They're graduating into a job market
00:11:02.980 that's been largely destroyed for two primary reasons. First of all, big tech companies,
00:11:08.500 including Google, are replacing American workers with H-1B visa holders. And secondly, of course,
00:11:13.920 AI is rendering many entry-level jobs obsolete. You know, 10 years ago, it was common for smart,
00:11:18.460 highly educated students from a school like University of Arizona to get a job making
00:11:22.540 Excel spreadsheets as a consultant or writing low-level code as a programmer or scanning
00:11:27.100 through stacks of legal documents as a paralegal or handling complaints from important clients as
00:11:32.640 a customer service rep. Today, AI can perform many of those roles, or at least it simplifies
00:11:39.640 many of those tasks, which means that fewer humans are needed to perform them, you know,
00:11:44.380 sort of like the self-checkout phenomenon, but happening at scale. Now, at the University of
00:11:49.280 Arizona, every single student is understandably terrified about this. Even the students who have
00:11:54.940 secured a good job are worried that in a few years, they'll be replaced too. When you're a
00:12:01.500 wealthy tech executive talking to an audience like that, it would help to address their concerns
00:12:05.680 directly. It's not as though the students don't recognize the potential benefits of AI.
00:12:10.480 Many of them probably used ChatGPT to help them with their assignments,
00:12:14.940 or more realistically, to do their assignments for them. Now, the issue is that from the student's
00:12:19.520 perspective, AI, even alongside its benefits, is going to prevent them from ever owning a home
00:12:28.840 or having a real career, or starting a family. Now, one way to reassure these students would
00:12:35.280 be to tell them the truth about what AI is and what it's capable of. No, AI is not conscious.
00:12:41.680 It has no sense of humor. It has no capacity for love or wisdom. It cannot articulate a single
00:12:47.680 original argument or insight on any topic. It's not capable of generating a unique creative piece
00:12:54.280 of content, whether we're talking about a script or a poem or a song or a film or a wedding toast.
00:12:59.900 Now, it could provide potentially worthwhile analysis and feedback on the creative things
00:13:04.300 that you make, but it can't make its own from scratch. You can't even trust that it's fact
00:13:11.540 checking properly since half the time it's hallucinating in order to give you the answer
00:13:14.840 that it thinks you want to hear. Even the leading AI models right now will tell you a lie to your
00:13:19.700 face without any hesitation or self-awareness because it has no capacity for self-awareness
00:13:23.940 because it's not a self. And that's why when you see AI art, it's always some bastardized
00:13:29.100 combination of existing art styles. It's never going to be anything new. And these have been
00:13:34.640 persistent problems for every AI chatbot for several years now. And no one's been able to
00:13:39.160 fix them. And that's because, as we'll discuss in some detail in a moment, all of these AI models
00:13:42.900 are simply gathering an enormous amount of information, most of it's stolen, and trying to
00:13:47.700 synthesize it in order to carry a conversation with you. And while the technology has some
00:13:54.180 important uses, it can never and will never replace the capacity of the human brain.
00:14:01.840 Eric Schmidt will never say any of that out loud because he has billions of dollars invested in
00:14:05.880 AI companies. His goal is to drive up the value of these investments as much as he can. He thinks
00:14:10.880 that there are no drawbacks to the strategy, but indeed there are. Now, it's very possible that
00:14:16.180 if AI billionaires keep talking like this, we'll have a total ban on AI in this country
00:14:22.020 in the future. And that would be an overreaction. It would damage our economy.
00:14:28.080 But it would also be inevitable because no one wants to live in a country that's dominated
00:14:32.680 totally by robots. Now, that's not to dump on Eric Schmidt too much. He's just one of many
00:14:38.740 examples. Here's Sam Altman, the founder of OpenAI. Altman began OpenAI as a nonprofit with Elon Musk
00:14:45.220 before realizing he could make a lot of money by taking all the donations he
00:14:48.660 received and running OpenAI as a private company, which it is now.
00:14:52.620 Watch.
00:14:54.340 Kid born today will never be smarter than AI.
00:14:57.660 Ever.
00:14:58.340 And a kid born today, by the time that kid like kind of understands the way the
00:15:03.180 world works, will just always be used to an incredibly fast rate of things
00:15:08.020 improving and discovering new science.
00:15:09.940 They'll just, they will never know any other world.
00:15:11.700 It will seem totally natural.
00:15:12.900 will seem unthinkable and Stone Age-like that we used to use computers or phones or any kind
00:15:18.800 of technology that was not way smarter than we were. A kid born today will never be smarter than
00:15:25.800 AI. Now, think about what this statement reveals. Sam Altman believes that intelligence is the sheer
00:15:32.740 amount of data that you have in your brain, the number of facts you can digest and recall at a
00:15:38.500 moment's notice. That is the sum total of your intellect, according to Sam Altman.
00:15:44.040 But actually, your intellect is a lot more complicated than that. Unlike AI, you are a
00:15:48.820 being with the capacity for contemplation and reflection and intuition and innovation.
00:15:54.940 You can come up with solutions to problems that would never occur to AI because they've never
00:16:00.360 occurred to anyone else in the history of humanity. And if they never occurred to anyone
00:16:03.500 else, then they can't occur to AI because AI is getting it from what everyone else has already
00:16:07.800 said. And for that reason, AI will always be intellectually inferior to human beings. Human
00:16:15.580 intelligence will always be, in this important sense, superior to artificial intelligence.
00:16:22.120 And that's because human intelligence is human. It's capable of a depth that AI can never reach.
00:16:28.200 Humans know what it's like to actually be, you know, to exist as individual selves in the world.
00:16:34.180 our greatest insights and works of art derive from this experience and self-understanding.
00:16:39.720 AI has no such experience and never will, which is why the insights it generates on its own
00:16:45.060 are always so hollow. For example, just to illustrate the point, I asked ChatGPT
00:16:50.940 to generate the best sad poem that it possibly can. I didn't give it any other prompts. I didn't
00:16:57.620 help it to formulate the poem. Just that prompt alone, give me a sad poem. Best you could do.
00:17:01.820 And here's what it spit out. I will recite it for you now. Here's what it said.
00:17:06.440 You left so quietly, the walls still ache, still holding shadows only you could make.
00:17:11.720 Your name survives in dust and window light, a ghost that visits every sleepless night.
00:17:16.940 I speak to silence like it answers back and memorize the sound of what I lack.
00:17:21.600 The cruelest part of losing you, I've found, is waking up each day without a sound.
00:17:26.700 now as a gimmick it's impressive that a computer can just vomit that out in three seconds
00:17:33.820 and this is also admittedly a better poem than what many humans could write it's probably better
00:17:39.920 what i could write i'm not a poet but it's it's not a good poem by any stretch of the imagination
00:17:46.580 it's hackneyed and cliche there's nothing interesting about it the visuals and metaphors
00:17:51.060 are all sort of standard issue which is what you would expect chat gpt is cribbing from every sad
00:17:56.400 poem ever written and coming up with what is basically the poetry equivalent of a statistical
00:18:01.340 average. Now, if an eighth grader wrote that as a creative writing assignment and actually wrote it,
00:18:08.200 didn't just get it from Chacham B, you'd give it a solid B+. If Amanda Gorman wrote that,
00:18:13.700 it would represent a major evolution in her craft. But if you were flipping through a book
00:18:18.320 of classic works of poetry and stumbled on that, you'd assume that somebody had inserted it as a
00:18:23.880 practical joke, because in that context, it would stick out like a sore thumb. So here's the
00:18:30.300 question. Just using poetry as an example, of course, not that it all comes down to this,
00:18:35.260 but if you are someone who can write a poem that, unlike that one I just performed,
00:18:42.140 is actually beautiful and unique and inventive and interesting and evocative and singular,
00:18:48.080 are you smarter than chat gpt i would say in a really meaningful way the answer is yes
00:18:55.580 now you cannot consciously store and regurgitate information like chat gpt then again computers
00:19:01.640 have been better at doing that for as long as computers have existed but you can meditate and
00:19:07.620 contemplate and create in a way that chat gpt cannot that is a certain kind of intelligence
00:19:13.540 I would say it's the most important kind. It's the kind that makes us human.
00:19:18.260 And it's what makes humans smarter. Some humans, at least.
00:19:24.300 Sam Altman is clearly in denial about this. Watch.
00:19:28.700 Fundamentally, our business, and I think the business of every other model provider, is going to look like selling tokens.
00:19:36.520 You know, they may come from bigger or smaller models, which makes them more or less expensive.
00:19:41.780 They may use more or less reasoning, which also makes them more or less expensive.
00:19:45.360 They may be running all the time in the background trying to help you out.
00:19:48.820 They may run only when you need them if you want to pay less.
00:19:51.700 They may work super hard, you know, spend tens of millions, hundreds of millions, someday billions of dollars on a single problem.
00:19:59.300 It's really valuable.
00:20:00.240 But we see a future where intelligence is a utility
00:20:08.260 Like electricity or water
00:20:10.460 And people buy it from us on a meter
00:20:14.840 And use it for whatever they want to use it for
00:20:17.080 The demand that we see for that
00:20:20.040 Seems like it's going to continue to just go like this
00:20:22.800 And if we don't have enough
00:20:27.180 We either can't sell it or the price gets really high and it kind of goes to rich people or society makes a bunch of sort of central planning decisions that I think almost always go badly about, you know, we're going to use our limited compute supply for this and not that.
00:20:43.320 So the best thing to me throughout all the history of capitalism, innovation, whatever you want, is to just flood the market.
00:20:49.640 Yeah.
00:20:49.960 now notice how he phrased that we see a future where intelligence is a utility like electricity
00:20:57.020 or water he doesn't say that artificial intelligence will be a utility he says that
00:21:01.200 intelligence in general will be treated like a utility we'll have to pay for it and there won't
00:21:06.980 be any competition just like your power company open ai will have a monopoly they'll send you a
00:21:12.140 bill every month and you'll be compelled to fork over the money in exchange for intelligence and
00:21:17.320 in case you're not convinced, Altman makes this claim while looking like a man who's just overdosed
00:21:21.260 on SSRIs. I mean, dystopian does not begin to describe what we just saw there. But the silver
00:21:29.640 lining is that Altman is just making things up and saying them out loud, which is what he often
00:21:34.620 does. Aside from the obvious financial motivations, he has a legal reason to make a statement like
00:21:39.540 this. Because you see, the way that AI companies train their products, the way that they get their
00:21:44.340 alleged intelligence is by throwing a massive amount of human-created information into the
00:21:49.520 model. They don't simply crawl the internet and lift information from news sites and WebMD and
00:21:54.280 social media and so on. They do that, but they also scan through books, lots and lots of books.
00:22:00.120 And when you do this kind of large-scale content theft, you run a very large risk of violating
00:22:04.880 copyright laws. Now, under our copyright laws, in most cases, it's legal to make a copy of somebody
00:22:10.640 else's creative work. But there's an exception called, illegal rather, to do that. But there's
00:22:15.600 an exception called fair use, which you probably heard of. If you're using a large amount of
00:22:22.000 content that was created by other people, then the best way to argue fair use is to claim that
00:22:27.240 you have transformed the content in some way, meaning you're using it for a completely different
00:22:31.860 purpose than it was originally intended for a different audience. The classic example
00:22:35.380 would be criticism or parody. For example, when I play a video on this show, as I just did,
00:22:40.640 I'm offering my commentary and that's considered transformative. Along the same line, Sam Altman
00:22:46.480 and OpenAI are claiming that they are transforming all of the content that they're taking. They're
00:22:51.600 admitting that they're copying millions of books and websites, but they're saying that they're
00:22:55.380 allowed to do that because they're putting those books and websites to a transformative use,
00:23:00.340 specifically training their AI models. This is why Sam Altman and Eric Schmidt and everybody else
00:23:07.060 feels the need to claim that AI is more powerful than it actually is. If they simply admitted that,
00:23:12.740 in fact, AI is just a gigantic amalgamation of content from other people, then, well,
00:23:18.580 the courts might shut them down overnight. They simply cannot make that acknowledgement
00:23:24.300 under any circumstances, or the entire industry could be destroyed. So instead, all these
00:23:29.740 executives are obligated to push the idea that they've created a new form of intelligence,
00:23:34.120 which is eventually going to replace human intelligence entirely.
00:23:38.700 Now, by any definition, if they pulled that off,
00:23:41.140 then they'd certainly have a transformative use on their hands for sure.
00:23:46.520 The co-founder of Anthropic, which makes the Claude AI, is making the same claim.
00:23:51.060 He just spoke at the Vatican to urge the church to assist all the poor people
00:23:54.380 who will never work again once AI takes over the economy.
00:23:58.600 He also suggested that AIs have developed human-like characteristics,
00:24:02.220 including introspection. Watch. There is a real possibility that AI will displace
00:24:08.460 human labor at a very large scale. If that happens, supporting those displaced will be
00:24:13.760 a moral imperative of historic proportions. This task will be difficult enough, but I worry most
00:24:20.100 dialogue misses an even harder challenge. AI development is concentrated in a handful of
00:24:25.860 wealthy nations. How will we ensure that the gains of AI are shared globally? We do not have a
00:24:31.400 mechanism for this. It is an unsolved problem. And it is the kind of problem the church has
00:24:37.000 historically refused to let the world ignore. And I will be honest, we keep finding things that are
00:24:42.580 mysterious, even unsettling. We find structures that mirror results from human neuroscience.
00:24:47.640 We find evidence of introspection. We find internal states that, functionally, mirror joy,
00:24:53.020 satisfaction, fear, grief, and unease. I don't know what that means, but I think it warrants
00:24:59.640 ongoing discernment.
00:25:03.540 So this has been a recurring theme in science fiction for a very long time.
00:25:08.580 Kurt Vonnegut wrote Epicac in the 1950s, for example, about a machine that falls in love
00:25:14.040 with a woman and then destroys itself when it realizes she'll never love him back.
00:25:18.080 And now all these tech executives want us to believe that they've made it a reality.
00:25:23.620 But Epicac would have been a much worse story if the humans had told the machine about the
00:25:28.520 concept of suicide and how to do it. And then the reader would say, oh, so the machine just
00:25:32.820 did what it was instructed to do. And in that case, all the deeper themes are lost. But
00:25:37.640 that's exactly what modern AIs are doing. Everything they do at one point or another
00:25:42.420 was spelled out by a person. The reality is a lot less dramatic than Sam Altman or Anthropic will
00:25:48.300 admit. A recent court case out of California involving Anthropic shed a lot of light on
00:25:54.040 exactly what these AI companies are doing. This is a quote from the judge's ruling, quote,
00:25:58.860 Anthropic spent many millions of dollars to purchase millions of print books, often in used
00:26:02.580 condition. Then its services providers stripped the books from their bindings, cut the pages to
00:26:09.000 size, and scanned the books into digital form, discarding the paper originals. Each print book
00:26:13.160 resulted in a PDF copy containing images of the scanned pages with machine-readable text,
00:26:18.420 including front and back cover scans for soft cover books. Anthropic created its own catalog
00:26:23.620 of bibliographic metadata for the books it was acquiring.
00:26:27.540 It acquired copies of millions of books.
00:26:30.360 So to recap, they bought millions of books
00:26:32.800 and totally destroyed them.
00:26:34.680 And all the people who claimed to care about book burning
00:26:37.000 were, of course, silent throughout this process.
00:26:39.780 And by the way, Anthropic also pirated a lot of books
00:26:41.640 from the internet as well,
00:26:42.680 although they say they stopped doing that
00:26:44.220 in favor of buying the books and ripping them apart.
00:26:46.760 But wait a minute, you might ask.
00:26:49.120 Now, what happens when Anthropic and Google and Meta
00:26:51.700 and OpenAI and everyone else
00:26:52.920 eventually runs out of books to scan?
00:26:56.620 What happens when they've mined
00:26:58.080 the entire internet for information?
00:27:00.160 How do the models continue to improve after that?
00:27:03.400 Well, it's a pretty important question.
00:27:04.880 And right now, as we all know,
00:27:06.000 these AI models are very far
00:27:07.460 from human level intelligence
00:27:08.720 or anything close to it.
00:27:10.160 Once they run out of your data,
00:27:12.220 how are they going to get any better?
00:27:15.120 Reese's knows a thing or two
00:27:16.480 about great combinations.
00:27:18.440 Chocolate and peanut butter, obviously,
00:27:20.500 but there's more than one way to Reese's.
00:27:22.920 From indulgent Reese's Big Cups with caramel to crunchy Reese's pieces and Reese's miniatures, there's a delicious Reese's for every mood.
00:27:31.320 It's the same combo you love, just with more ways to enjoy it.
00:27:34.940 So whether you're snacking, sharing, or just treating yourself, nothing else is Reese's.
00:27:42.600 That's a question you're really not supposed to ask.
00:27:44.620 And the answer is that big tech wants you to spend your life adding new contributions for the benefit of their models.
00:27:51.760 They want you to work for them for free.
00:27:53.860 In other words, every time you ask their AI a question or tell their AI something, you're improving the model.
00:27:59.000 Every time you offer an original thought on social media or upload a photograph or a drawing, you're improving the model.
00:28:05.660 Anytime you put any content on the Internet at all, as we're doing right now, you're improving the model.
00:28:10.500 And more importantly, every dollar you earn, no matter what your job may be, must contribute towards the constant growth of new AI data centers to keep the churn going.
00:28:18.520 BlackRock CEO Larry Fink just came out and said this. Watch.
00:28:24.500 Americans, to think about growing with the United States, we will have far than enough money to invest in this infrastructure.
00:28:36.280 But as the governor was talking about, the need for electrons is growing every day.
00:28:42.380 some of these
00:28:45.280 if we're going to be the leader
00:28:47.800 in technology, which we are
00:28:49.540 if we are going to be the leader
00:28:51.680 in AI, which we presently
00:28:54.040 are, it's just going to
00:28:55.900 require trillions of dollars of investments
00:28:58.180 and if we don't invest 0.99
00:29:00.120 in it, China will be the
00:29:01.860 global leader in this
00:29:03.000 and so to me it's not whether
00:29:05.480 this is a must
00:29:07.860 and if you
00:29:11.900 think about how that translates it translates into a more dynamic economy we need the united
00:29:19.340 states economy to grow it over over two percent we need the u.s economy to grow at three percent
00:29:24.360 especially with the growing deficits the federal government has and so much of this money not just
00:29:31.360 the project is going to be coming from the private sector from savings accounts from pension accounts
00:29:36.720 from insurance companies and on and on and on.
00:29:39.780 Similar to the ESG scam, which BlackRock was also heavily involved in,
00:29:43.740 the big funds and banks take assets that people have earmarked
00:29:46.500 for their 401ks or savings accounts,
00:29:48.780 and they invest that money in their pet projects.
00:29:51.680 A few years ago, these pet projects involved solar panels and reusable farts,
00:29:56.060 and today those projects involve data centers.
00:29:59.440 And you really can't opt out unless you completely abandon the big funds entirely,
00:30:03.060 which for most people with a 401k means you have to pay a lot more in taxes.
00:30:07.240 As it stands, if you contribute to your 401k,
00:30:10.140 a large chunk of that contribution is going to companies that are building
00:30:13.040 or investing in data centers, whether you like it or not.
00:30:17.080 If you pay attention, it's not hard to realize that the AI boosters
00:30:19.760 are overstating their case.
00:30:21.060 Here's Anthropic CEO lowering expectations just a tad
00:30:24.060 in response to a question from a journalist.
00:30:27.060 Watch.
00:30:28.380 Give me one or two examples of what could go wrong.
00:30:32.500 So the kind of thing that we, there are two classes of things that I can imagine could go wrong.
00:30:38.480 One, again, is around this idea of reliability, which is just it targets the wrong person, it shoots a civilian,
00:30:46.180 it doesn't show the judgment that a human soldier would show.
00:30:52.060 A friendly fire, or shooting a civilian, or just the wrong kind of things.
00:30:56.880 We don't want to sell something that we don't think is reliable,
00:30:59.920 and we don't want to sell something
00:31:01.420 that could get our own people killed
00:31:02.700 or that could get innocent people killed.
00:31:07.900 You did mention shooting civilians
00:31:09.500 like two or three times there,
00:31:11.000 which is, it's a little creepy
00:31:13.760 that he was so focused on that.
00:31:16.600 You know, there's a few things
00:31:17.420 that could go wrong, you know, 0.98
00:31:18.400 shooting civilians, shooting civilians. 0.92
00:31:22.620 This thing's really going to start 0.99
00:31:23.720 shooting civilians soon, 0.85
00:31:24.800 just so you guys know. 0.95
00:31:26.920 It's good that he doesn't want to sell
00:31:28.360 a product that will murder American civilians or soldiers. And the question is, how exactly can he
00:31:33.160 make that assurance? If the AI is trained on millions of books and blog posts and makes
00:31:37.400 decisions based on that data, then no human can really predict with complete certainty how the AI
00:31:41.760 will behave in a novel life or death situation. I mean, AI can probably handle basic scenarios like
00:31:46.980 driving a car down a simple road or making sure a rocket hits a valid target. These are scenarios
00:31:52.860 where there's plenty of training data. But what happens when there's no training data?
00:31:56.740 So already in very simple scenarios, AI is repeatedly failing and humiliating anyone who's dumb or trusting enough to totally rely on it.
00:32:05.920 Recently, a prosecutor in Georgia had to explain to the state Supreme Court why she relied on AI to create an important filing, which was riddled with fake case citations.
00:32:15.820 Watch.
00:32:17.080 If there are no more questions.
00:32:19.200 So before you sit down, there's one more thing I need to ask you about, unfortunately.
00:32:23.300 In reviewing the trial court's order denying the motion for new trial, there are at least
00:32:31.400 five citations to cases that don't exist, and there's at least five more citations to
00:32:37.740 cases that do not support the proposition for which they're cited, including three quotations
00:32:44.860 that don't exist.
00:32:46.820 My understanding is that you prepared the denial order for the trial court.
00:32:54.740 Were those citations in the version of the order that you submitted to the trial court?
00:32:59.320 No, Your Honor. I do not believe so. They were not.
00:33:01.940 I did prepare an order. That order was revised.
00:33:07.940 So those nonexistent cases were cited in your initial brief opposing the motion for new trial?
00:33:15.820 Your honor, I'm not aware of that, but I would be glad to research that.
00:33:40.280 that's a prosecutor in that clip someone who's responsible for sending people to prison for a 0.62
00:33:47.240 long time and her office is relying on fake cases created by an ai out of thin air you have to
00:33:53.000 wonder how often this kind of thing happens without being detected the big law firm sullivan and
00:33:56.940 cromwell one of the most famous and high-priced law and for law firms in the world was just caught
00:34:01.500 doing the same thing using ai to deceive the court quoting from the new york times uh an elite wall
00:34:06.560 Wall Street Journal, rather elite Wall Street law firm, has apologized to a federal judge for
00:34:12.560 submitting a court filing replete with errors created by artificial intelligence, including
00:34:17.040 hallucinations that fabricated case citations. The AI-generated errors came in a recent motion
00:34:23.180 in U.S. Bankruptcy Court in Manhattan. The firm provided a ledger of the errors, which spanned
00:34:27.780 three pages and totaled around three dozen. A number of them involved the citation of seemingly
00:34:33.020 imagine passages from real cases. So, again, it's happening all over the country. This is
00:34:37.840 in Minnesota. Watch. In a recent court filing in Hennepin County Court,
00:34:43.240 Attorney Frederick Knack wrote that prior Minnesota cases support his argument,
00:34:48.040 citing a 1992 case called State by Sundquist versus Provost. The problem? That case doesn't
00:34:54.540 exist. Neither did the other case Knack cited right before it or another case cited later.
00:35:00.220 When Judge Lori Miller caught it, she wrote the court wonders if this citation may be the result of an AI-generated hallucination.
00:35:07.700 It's something that created to give you a convincing answer of what you asked for, but it's not necessarily an accurate answer.
00:35:13.700 David Larson is a law professor at Mitchell Hamlin teaching a course next semester on artificial intelligence and the law.
00:35:19.520 Is this happening everywhere?
00:35:21.260 I think it's fair to say it's happening everywhere.
00:35:23.280 One thing I believe is that the temptation to use AI is so strong that people just can't resist it.
00:35:31.320 According to a database tracking AI hallucinations in court filings,
00:35:34.780 there are 134 cases across the country in which an attorney cited fabricated case law.
00:35:40.760 Carol Levin News found one other case in Minnesota.
00:35:43.120 In July, when Judge Christian Sandy caught attorney David Lutz citing a phony case,
00:35:48.200 Lutz confessed, admitting he used AI and forgot to cross-reference the cases.
00:35:52.960 all these lawyers should be arrested and disbarred immediately but that's not happening uh the
00:36:00.660 georgia da got a slap on the wrist and as you just heard these minnesota attorneys simply have to pay
00:36:05.080 a fine and maybe take a scolding from the state bar it's enough to make you wonder when we're
00:36:11.260 going to get full-fledged ai attorneys who make a complete mockery of the courtroom and get every
00:36:15.820 case wrong um the more sam altman tells us that ai super intelligence has arrived the more tempted
00:36:21.640 these attorneys will be. Actually, in one case in New York, this has already happened. An elderly man
00:36:26.500 representing himself deployed an AI to speak to the court on his behalf, and it didn't really go
00:36:33.460 well much. May it please the court. I come here today a humble proceed before a panel of five
00:36:40.360 distinguished justices. Is this, hold on. Is that counsel for the case? I generated that.
00:36:50.480 I'm sorry?
00:36:51.620 I generated that?
00:36:53.400 That is not a real person.
00:36:58.240 Okay.
00:36:59.020 It would have been nice to know that when you made your application.
00:37:03.780 You did not tell me that, sir.
00:37:05.860 I received the application.
00:37:07.260 And you have appeared before this court and been able to testify verbally in the past.
00:37:14.020 You have gone to my clerk's office and held verbal conversations with our staff for over 30 minutes.
00:37:21.340 Okay?
00:37:21.880 I don't appreciate being misled.
00:37:26.760 So either you are suffering from an ailment that prevents you from being able to articulate, or you don't.
00:37:35.440 You are not going to use this courtroom as a launch for your business, sir.
00:37:42.080 So if you are able to shut that off.
00:37:50.760 Well, the AI lawyer is having a good time.
00:37:52.780 You can tell that.
00:37:54.000 It's safe to say the AI did not, in fact, please the court.
00:37:57.860 And yet on and on it goes across every industry.
00:38:00.420 This is from The Atlantic in an article published just the other day.
00:38:03.140 Quote, earlier this week, the New York Times reported that the future of truth,
00:38:07.480 Stephen Rosenbaum's much-discussed book about how AI shapes reality contains more than half a dozen fake or misattributed quotes.
00:38:14.880 Rosenbaum pinned some of them on his use of AI.
00:38:18.340 He claimed responsibility for the errors, said he was investigating what went wrong.
00:38:22.560 By the time I spoke with him on Thursday, though, he was pointing his finger elsewhere.
00:38:27.280 ChatGPT effed up the book, Rosenbaum said.
00:38:29.660 It's been a rough week for human authorship all around.
00:38:32.060 On Monday, a viral post showed a Nobel-winning novelist seemingly admitting to using AI to sharpen her story ideas
00:38:37.860 before later claiming she had been misunderstood.
00:38:40.620 On Tuesday, allegations mounted that the Trinidadian author Jameer Nazir had used AI to write The Serpent in the Grove,
00:38:47.920 which won the Commonwealth Short Story Prize.
00:38:50.660 By Wednesday, two of the other five prize winners had come under similar scrutiny.
00:38:55.700 Now, seeing all these cases, you have to ask yourself, when exactly are we getting these superhuman intelligence from these AI chatbots?
00:39:04.720 They can't write novels or legal briefs without failing in spectacular fashion and destroying careers in the process.
00:39:11.960 As I've always said, the technology is extremely impressive and very useful in many ways, but there's a clear ceiling to the knowledge base of these AI chatbots.
00:39:21.940 There's only so many books and websites you can rip off before the AI has to start thinking for itself and before people stop voluntarily training these models for free.
00:39:31.360 If every leading AI is already failing in significant ways, then how much improvement can we really expect exactly?
00:39:38.620 There are entire industries built around convincing people to massively overpay for things that they barely think about.
00:39:44.420 Cell phone service is one of the best examples. People will spend $90 a month for wireless service and never even question it.
00:39:50.420 The bill just arrives every month like a tax, and the big wireless companies keep raising prices because they assume nobody's going to leave anyway.
00:39:58.280 That's why Pure Talk makes so much sense.
00:39:59.860 Right now, they're giving you unlimited high-speed data for just $34.99 a month.
00:40:04.140 What's even crazier is that unlimited high-speed data at Pure Talk used to start at $55 a month,
00:40:08.860 but instead of constantly finding new ways to charge customers more, Pure Talk keeps pushing to give people more for less.
00:40:14.940 So if you checked out Pure Talk before and never made the switch, it's worth taking another look.
00:40:20.120 And if you're skeptical because you assume, you know, cheaper automatically means worse, well, try it for yourself for 30 days.
00:40:26.620 No contract, no cancellation fees, no risk.
00:40:29.040 With PureTalk, you can switch in as little as 10 minutes.
00:40:31.260 And if you need help, you can actually talk to a real U.S.-based customer service team, not an automated robot voice asking you to press 700 times before disconnecting the call.
00:40:41.600 Go to puretalk.com slash Walsh to claim unlimited high-speed data for just $34.99.
00:40:46.480 Again, that's puretalk.com slash Walsh to switch to my wireless company, America's wireless company, Pure Talk.
00:40:52.980 A lot of universities still operate like it's 1978.
00:40:55.940 Same building, same approach, same idea that, you know, students should spend four years in a mountain of money getting a degree that may or may not prepare them for anything.
00:41:04.680 Meanwhile, the real world changes every six months.
00:41:06.660 And that's one reason Grand Canyon University, a private nonprofit Christian university based in Phoenix, Arizona, stands out.
00:41:12.660 My show is proud to be supported by a university that's focused on growth and relevance instead of just protecting the tradition for the sake of the tradition.
00:41:20.340 75% of GCU's programs and facilities have been built in just the last 10 years.
00:41:24.100 That means modern classrooms, modern tech, and programs designed around the careers that actually exist now, not 20 years ago.
00:41:31.540 GCU now offers hundreds of programs overall, including degree programs, emphases, and certificate programs.
00:41:39.200 And despite being one of the largest universities in the country, they've kept tuition frozen on the traditional campus for the past 17 years,
00:41:45.460 which is almost impossible to believe in modern higher education.
00:41:48.780 But it's true, especially when most colleges raise prices every chance they get.
00:41:54.480 GCU also awarded more than $400 million in institutional scholarships in 2025 alone.
00:41:59.480 Their campus was also ranked among the top 20 college campuses in the country.
00:42:03.880 And this is a real major university.
00:42:06.320 More than 132,000 students attend GCU, either online or on campus.
00:42:11.360 Their basketball program has already made multiple NCAA tournament appearances since joining Division I.
00:42:17.160 Now they compete in the Mountain West Conference.
00:42:20.000 Grounded in Christian values, GCU is focused on preparing students to lead with integrity,
00:42:25.100 serve with purpose, and build meaningful careers and communities.
00:42:28.320 Find your purpose at GCU, private, Christian, affordable, nonprofit.
00:42:32.880 Visit gcu.edu to learn more.
00:42:36.320 Now, as I mentioned, we discussed AI data centers a couple weeks ago.
00:42:40.360 I'm not suggesting that AI has no valuable uses or that it can't improve.
00:42:44.860 AI can obviously benefit the country.
00:42:47.260 It can launch businesses and perform valuable services.
00:42:49.680 It can give certain kinds of useful feedback.
00:42:51.960 It can even, I mean, it can diagnose medical conditions,
00:42:55.160 sometimes more reliably than many human doctors who are also using AI anyway.
00:43:00.380 It can save lives, potentially, given the right circumstances.
00:43:04.240 But if we keep pretending that AI has human-like capabilities, it never actually has.
00:43:10.680 If we keep threatening to destroy millions of jobs and prevent young people from even starting a life of their own,
00:43:17.080 then before long, no one will be able to use AI for any reason.
00:43:21.440 There will be a revolt that forecloses the possibility and probably does a lot of damage in the process.
00:43:27.300 It's what Ted Kaczynski wrote about, sitting alone in his cabin in Montana.
00:43:30.460 Right now, ironically enough, the people who are doing the most to bring down the industrial technological system he was talking about are the leaders of the industrial technological system themselves.
00:43:42.600 It's not activists. It's not students on college campuses.
00:43:46.260 The more that tech billionaires like Sam Altman and Eric Schmidt open their mouths, the more they push us toward the primitive state of wild nature that Ted Kaczynski called for about three decades ago.
00:43:57.300 We can either rein in the gloating and articulate some realistic guardrails on where this technology is heading, or we can devolve into precisely the primitive dystopia that the Unabomber dreamed of.
00:44:14.840 That'll do it for the show today. Thanks for watching. Thanks for listening. Talk to you tomorrow. Have a great day. Godspeed.
00:44:19.360 Martin Luther King Jr. is an American icon
00:44:28.020 Widely considered one of the greatest Americans who ever lived
00:44:30.880 A man who had a vision for a colorblind society 0.74
00:44:34.640 A post-racial America 0.75
00:44:36.240 He had a dream 0.94
00:44:38.180 It's just not the dream you thought it was
00:44:40.540 Were his true aims a colorblind society
00:44:43.420 Or something far more radical
00:44:45.080 Who bankrolled him
00:44:46.720 What unfolded behind the scenes in Birmingham, Alabama in 1963?
00:44:51.600 Was civil disobedience actually peaceful?
00:44:56.200 We wanted to show you a clip of the I Have a Dream speech,
00:44:59.300 but according to our lawyers, we can't.
00:45:01.180 In fact, King's family has made a lot of money
00:45:03.300 suing media outlets.
00:45:04.500 They want to silence critics like us.
00:45:07.000 What they're doing makes it very difficult
00:45:08.520 to judge Martin Luther King Jr.
00:45:10.180 not by the color of his skin,
00:45:12.040 but by the content of his character.
00:45:14.180 Is America today stronger, more unified,
00:45:17.260 and racially equal than before King's rise?
00:45:20.320 These questions demand answers, and as Americans,
00:45:22.700 we are entitled to a full accounting
00:45:24.500 of the Civil Rights Movement and its consequences.
00:45:26.760 King's Movement fundamentally transformed our country
00:45:29.640 and our system of government.
00:45:31.080 I speak as a citizen of the world.
00:45:33.940 Each day the war goes on, the hatred increases,
00:45:37.920 though the cause of evil prosper.
00:45:40.560 First part of our two-part special
00:45:42.160 on the Civil Rights Movement,
00:45:43.480 A new constitution. Available now on Daily Wire Plus.
00:45:48.760 Before you knew what a stock was, you traded snacks, cards, turns.
00:45:54.500 You knew what something was worth because you felt it.
00:45:57.960 That instinct to trade didn't disappear. It just grew up.
00:46:02.160 TD Easy Trade taps into that instinct so you can build something real for the future.
00:46:06.780 With no minimums, no monthly fees, and 100 free trades.
00:46:10.480 you already got this
00:46:12.180 because you are made to trade
00:46:14.700 and TD Easy Trade is made to help
00:46:17.100 download it now