Skip to content
PodSubs
ShowsPeopleGuestsJobsNewsEvents
More
BlogPodCred
Menu
ShowsPeopleGuestsJobsNewsEventsBlogPodCredNetworksSign in
Join nowSign in
  • Shows
  • People
  • Jobs
  • News
  • More
    • Guests
    • Events
    • Blog
    • PodCred
    • Networks
    Join nowSign in

ReadMultiplex.com Podcast.

ReadMultiplex.com Podcast. artwork

ReadMultiplex.com: The “AI Detector” Grift Game: Why They Have Dumb Down Human Writing So As To Not Get “AI Flagged”.

Aug 24, 202636 minAudio

Got something to say to this podcast? Send a voicemail, and you might hear yourself on the next episode! The entire industry of text based “AI Detectors” is a pure grift machine built on arrogance, bad science, and the willfull stigmatizing of millions of real human writers. These tools do not work. They never worked. They will never work in any reliable way that matters. And the people selling them know it. I wrote a while ago about the use of Ai in any work of text in detail. I predicted some of these “AI Detector” companies would begin and it has. they are desperate for business through any means necessary.. Of course people are using AI, and calculators, and cars instead of walking and electric lights instead of candels. It is up to the author to disclose if they used such tools but these companies claim to solve a problem that does not exist in the way they pretend. The only questions that actually matter when you read something are simple: Is this passage worth reading? Is it thoughtful about the subject? Did I learn anything? An average human can answer those questions without some black box software that spits out a fake percentage and then ruins a student’s semester or a writer’s reputation. Detectors solve nothing. They create a new form of digital scarlet letter based on statistical noise. Read more at: ReadMultiplex.com Support this work by buying us a coffee: https://ko-fi.com/brianroemmele

Audio and video play straight from the show's own hosting. PodSubs stores and serves no media.

Transcript

From the show. Tap a time to play from there.

I want you to imagine a scenario for a second and really just try to put yourself in this person's shoes. Okay. You're a college senior, right? You have just spent the last three months meticulously researching, outlining, and crafting your final capstone essay. Yeah, that's a lot of work. Right, or... Or maybe you're like a mid -level manager at a logistics firm and you've spent weeks agonizing over this crucial market analysis report for the executive board. Oh, man. Yeah. The pressure.

Exactly. You poured over the primary sources. You refined every single sentence. You checked your transitions. You polished the vocabulary. You made absolutely sure the logical structure was completely flawless. And there's that singular moment of relief, right? You hit submit, you lean back in your chair, and you feel that specific rush of pride that only comes from a job genuinely well done. Yes. But the relief just doesn't last because almost instantly a piece of black box software, a program you have never seen, right?

A program whose internal rules you are literally not allowed to know. It scans your months of hard work and slaps a giant neon red completely fake label across the top of it. Oh, it's just it's devastating. It calls you a fraud to your professor or to your boss. And in a matter of seconds, your grade is ruined. Your professional reputation is in tatters. And worst of all. There is no human being you can appeal to. The machine has spoken. It is a genuinely terrifying position to be in. The psychological toll of that, just

being called a liar by an algorithm and having no way to defend yourself, it's devastating. And I think we need to be clear right from the start here, this is not some dystopian science fiction hypothetical. It's happening right now, this very morning. High schools, universities, and corporate HR departments all over the world. Which is exactly why we are dedicating this entire episode to unpacking this nightmare. Welcome to this deep dive for the ReadMultiplex .com podcast. Our mission today is to really zero in on a critical, incredibly urgent issue that

has been brought to light by the independent voice of Brian Aremmel. Yeah, his insights on this have been incredible. They really have. And I got to say up front, you know, in a tech landscape that is just flooded with venture capital hype and PR spin, it takes someone with Brian's candor and absolute clarity to cut through the industry noise. Absolutely. Because the core thesis we are exploring today, based entirely on... his insights, is explosive. We are looking at a system that is essentially a modern day digital witch trial. The entire industry of text

based AI detectors is a complete sham. And we really don't use the word sham lightly here. This isn't just a critique of a software bug or a glitch. We are talking about a systemic. foundational failure that is causing real measurable harm to innocent people. Massive harm. And we're going to look at the hard data today that proves these detectors are fundamentally mechanically broken. And furthermore, even the tools that occasionally seemed, I don't know, sort of accurate a year or two ago, they are rapidly becoming obsolete and wildly wrong. Yeah, the degradation

is real. So if you are listening to this right now. Whether you are a student writing a term paper, a professional drafting a proposal, or honestly just someone who sends carefully worded emails to clients, you need to pay close attention. This digital scarlet letter could absolutely affect you tomorrow. You know, what stands out to me in the source material is how... Incredibly fast, society just adopted these policing tools. Oh, overnight. Right. Without even stopping to ask the most basic fundamental question, which

is, do they actually work? We just sort of culturally accepted the premise. Figured, well, if artificial intelligence is smart enough to write human -sounding text, then surely a different artificial intelligence must be smart enough to detect it. Right. It sounds logical on the surface. It does. But to understand why these detectors fail so spectacularly, we have to look at their absolute foundation. We really have to look at what the machine considers to be human in the first place. OK, I want to stop you right there because this is where I get tripped up. When I think of a baseline for human writing, I think of, you know, the great

libraries of the world. I assume these massive tech companies train their models on encyclopedias, classic literature, academic journals, maybe the New York Times archives. So if that's the baseline, how are they getting it so wrong? I can completely understand why you would assume that. It sounds like the right way to do it. But it ignores the sheer scale of how large language models or LLMs are actually built. Okay. They didn't just read the classics. They essentially hoovered up the entire open Internet. Oh. And

if you think about the Internet. by sheer volume of text, it is not a curated library of high literature. The vast overwhelming majority of human text online is just casual, low -effort writing. Oh, I see where this is going. We're talking about comment sections, aren't we? We are talking about the sewage standard. The training data is dominated by Reddit posts, furious late night forum rants, sloppy personal blogs, social media comments, and quick text messages. It is that endless stream of consciousness, yo bro, casual style of communication. Because this disorganized

internet sludge makes up the statistical bulk of the training data, these systems now treat that chaotic style as the mathematical gold standard of what a human sounds like. Wait. Let me make sure I'm wrapping my head around this. Because there is so much literal garbage writing on the Internet, the detector's algorithm assumes that if a real -life human wrote a document, it simply must read like garbage. In a statistical sense, that is exactly how the math works. That is insane. It is. The detectors are scanning for something they call perplexity and random burstiness. They

expect human writing to be erratic. They want to see a total lack of structural predictability. Okay, so they want mistakes. Exactly. They are looking for a sentence that runs on way too long with terrible comma usage, immediately followed by a three -word fragment. They expect the vocabulary to vary wildly in a completely disorganized way. Therefore, if the algorithm encounters anything that is highly formal very careful rigorously structured or just uses a consistent professional tone it immediately flags it as suspicious that is just it's the tragedy of good writing you

are literally telling me that a student or professional is being actively penalized by the software for being excellent at their job the better you write the more you look like a machine to these detectors. Wow. And this is not just a theoretical edge case, right? The sources highlight that this flaw is baked into the very DNA of the product. If you feed these detectors highly vetted, beautifully written human text from long before ChatGPT or any LLM ever existed, the results are completely absurd. Like what? Passages from the Bible, for

example, regularly score as heavily AI generated. Wow. The Quran gets flagged constantly. The U .S. Constitution has been flagged over and over again as being written by a machine. Wait, the U .S. Constitution? Yeah. So the detector looks at the foundational document of the United States and concludes James Madison was basically using an early beta of generative AI. That is the conclusion the math forces it to make. The algorithm scans peak human writing from the 1950s. It scans Shakespeare. It scans formal academic papers written decades before modern AI was even a concept. All of it

gets marked as AI. That is wild. There is a famous example highlighted in the sources where the tool ZeroGPT scored the Book of Genesis at roughly 88 % AI. 88 % AI for the literal Book of Genesis. Yeah. I'm trying to think of an analogy here. It's like it's like grading a Michelin star chef on a curve based entirely on fast food drive through menus. Right. Exactly. The software is judging a perfectly plated duck confit and saying, well, there's no processed cheese on this and it doesn't come in a greasy paper bag. So a human

chef definitely didn't make it. It must be synthetic. If you write too well, if you use classic structure and repetition for emphasis, you are. penalized for not sounding like a 2 a .m. rant on a gaming forum. That Michelin star analogy hits the nail on the head. Think about the book of Genesis or really any classic literature or ancient religious text. It is highly formal. It uses deliberate repetitive structures for emphasis and oral memorization. Right. The begats and all that. Right. It is incredibly low in that random burstiness the

detectors are mathematically trained to expect from modern Internet sludge. So the algorithm sees this clean, structured, purposeful writing and computes. A human couldn't possibly be this organized and disciplined. This must be a machine. I have to ask, how do the companies that sell these detectors respond to that? Because if I'm a tech CEO, it is obviously deeply humiliating to have my flagship product claim the Bible was written by a Silicon Valley chatbot. Do they just deny it? The industry response is where the deep cynicism of this whole enterprise really

starts to show. To avoid looking completely ridiculous in the press, these companies have had to go into their software and manually add exceptions to their code. No. They essentially whitelist common, famous texts. They have to tell the software, hey, if you see the Declaration of Independence, just ignore your own mathematical analysis and label it human. Just to save face. Exactly. So they don't get caught making those specific, highly viral, embarrassing claims. I feel like I'm taking crazy pills here. If you have a core

algorithm and you have to manually tell the software, no, no, ignore your own analysis on this specific book because the results make our company look like a joke, doesn't that inherently prove the underlying algorithm is entirely flawed? You are admitting the math doesn't work. It proves exactly that. Their baseline is fundamentally broken. The exceptions list has to keep growing quietly in the background just so they can maintain the illusion of accuracy. And that illusion is what allows them to continue farming millions of dollars in licensing fees from schools and

universities. They are selling a surefire AI solution to terrified administrators, sometimes even getting their tools reinforced by state educational mandates. But the foundation is entirely makeshift. It's putting a flimsy digital band -aid on a gaping foundational wound. I hear what you're saying. And the Bible example is hilarious and damning. But I can easily imagine like a skeptical school board administrator listening to this and saying, sure, maybe the software gets confused by ancient archaic text because nobody talks like King James anymore. But we

aren't grading the Bible. We're grading modern everyday high school essays. Surely it works for normal 21st century writing. Tell me we don't just have to rely on anecdotal examples of old books. We don't have to rely on anecdotes at all. The source material brings the receipts, and the hard science has proven this failure across the board on modern text. Multiple independent peer -reviewed studies paint the exact same ugly picture. Let's look at the Weber -Wolfe study as a starting point. They tested 14 different AI detectors on the market. Crucially, that included

major, highly funded players like... Turnitin, which is absolutely ubiquitous in high schools and colleges right now, along with various free tools available online. OK, 14 different tools tested against modern writing. Let me guess. They caught the AI, but they flagged a bunch of the humans as well because they didn't write like Internet sludge. It was a failure on both sides of the equation. Every single one of the 14 tools scored below 80 percent accuracy. Below 80%. I want to pause on that number because in a casual mobile game, 80 % is fine. But in a

high stakes environment like education or professional academic publishing, a 20 plus percent failure rate is catastrophic. If a teacher has a class of 30 kids, this software is potentially giving false readings on six of them every single time an assignment is turned in. And the nature of those failures was incredibly erratic. Most of the tools lean heavily toward calling actual AI text. Wait, really? Yeah. Which means they fail at their primary advertised job of catching cheaters. But, simultaneously, they produced

enough false positives calling real, modern, human -writing AI to completely destroy any rational trust in the system. It's just a mess. It is. And what's worse, the researchers found a massive vulnerability regarding language. If a text was simply paraphrased, or if it was translated from another language by a non -native English speaker, the accuracy of these detectors tanked even further. further. Oh, man. Yeah. The tools were deemed completely inaccurate and unreliable for any serious application. Non -native speakers being

penalized more heavily, that adds a whole new layer of discrimination to this. But the absolute nail in the coffin, the study that just blew my mind when I was reading through Brian Rommel's insights in these sources, is the 2026 University of Florida study. Yes, the trainer study. Yeah. If you want to see a masterclass in exposing flawed technology, this is it. I want to spend some real time on this because the setup of this experiment is just brilliant. This was led by Patrick Traynor and his team, Seth Layton, Bernardo Medeiros, and Kevin Butler at the 2026 IE Symposium on Security and Privacy. Break down exactly how

they constructed this test because they eliminated any room for the detector companies to make excuses. They executed a very clever, very brutal methodology. The biggest defense detector companies use is claiming their test data was somehow tainted. Right. So Trainor's team eliminated that possibility. They went into the archives and gathered about 6 ,000 real, highly technical research papers that had been submitted to top -tier security conferences. And here is the crucial part. Every single one of these 6 ,000 papers was submitted

and published before JAT -GPT or any modern generative AI even existed. So they are 100 % verifiably human. There's absolutely zero question about their provenance. Zero question. They're the ultimate control group. Then the researchers used large language models to generate exact AI clones of those 6 ,000 papers. They fed the topics and structures into the AI and had it generate synthetic versions. So now they have 6 ,000 guaranteed human papers and 6 ,000 guaranteed AI clones of the exact same subject matter. They

took this massive data set and ran both sets through the five most popular, highly funded commercial AI text detectors on the market. This is the ultimate stress test. What happened when they tested the gold standard detectors against this perfectly clean data set? The results were an unmitigated disaster for the detector industry. Let's look at the false positives first. Again, a false positive is when the detector accuses a real human of being a machine. The false positive rates across the tools range from 0 .05 % all

the way up to a staggering 68 .6%. I need to make sure I heard that right. 68 .6 percent. You are saying one of the top commercial pools on the market looked at a database of guaranteed, verified human writing and accused the human of cheating nearly 70 percent of the time. That is exactly what the data shows. Over two thirds of the time, this highly expensive software labeled. brilliant, hardworking human researchers as machines. That is, I can't even process that. But if you think that's bad, the false negatives are where

the illusion completely shatters. A false negative is when the tool looks at actual 100 % AI generated text and confidently declares, yep, a human wrote this. The false negative rate swung from 0 .3 % to 99 .6%. 99 .6 % false negative. Let me visualize that for a second. That means if a university professor suspects cheating, Yep. It is doing

the exact opposite of what the sales brochure promises. It's essentially a random number generator at that point. It's statistical noise. It is arguably worse than a coin flip. Because a coin flip doesn't cost a university district millions of dollars in licensing fees, the detectors entirely collapsed under rigorous, controlled scientific testing. They became largely useless, completely unable to reliably tell the difference between human and machine. Patrick Treanor's conclusion was definitive and damning. What did he say?

He stated that these tools are not reliable, they are not robust enough to use, and likely never will be due to the underlying mechanics of how language models work. He specifically noted that commercially available detectors are poorly suited for deployment in academic or high stakes contexts. Think about the real world impact of a school ignoring that warning. We aren't just talking about abstract numbers on a spreadsheet here. We are talking about human lives, people's careers, their academic futures, their scholarships, their standing in their community. All of it is on the line based on tools that have entirely

collapsed under scientific scrutiny. They literally cannot tell the difference anymore. And yet universities, high schools and academic journals just keep buying the fantasy. They keep renewing the software licenses. Which naturally leads to the most important question we can ask about this entire situation. Why? If the hard science proves unequivocally that they do not work, why are they suddenly ubiquitous? Seriously, if they are mathematically proven to be a sham, if the trainer study exposed them this thoroughly, why is my local school

district suddenly mandating them on every single assignment? To understand that massive disconnect between reality and adoption, we have to look at the financial motivation. We have to examine what the sources call the grift cycle. And the most illuminating place to start is with the very company that kicked off the generative AI boom in the first place, OpenAI. The creators of ChatGPT, I vaguely remember this, didn't they try to make their own detector a couple of years ago? They did. OpenAI built a classifier tool specifically designed to detect AI -generated

text. They had the best engineers in the world, unlimited compute power, and intimate knowledge of how the models were built. But here is the reality check. They tested their own detector, and they realized it had a 9 % false positive rate, meaning it falsely accused innocent humans nearly 1 in 10 times. Simultaneously, it missed a vast fraction of actual AI text. And do you know what OpenAI did when they saw those numbers? I would hope they went back to the drawing board, but knowing tech companies, they probably shipped

it in beta. No, this is the fascinating part. They shut it down completely. Really? Yeah. Ethically, they looked at the data, they looked at the harm a 9 % false accusation rate would cause in schools, and they realized the accuracy was garbage. The very company that built the underlying models, the people who understand this specific technology better than anyone on earth, publicly admitted they could not make a detector that actually works safely. Which really should have been the end of the entire conversation. If the literal creators of the technology say, hey, guys, based

on the math, this is impossible to do reliably, everyone should have packed it up. But instead of listening to the engineers who built the AI, a massive cottage industry just sprang up out of nowhere to sell the fantasy anyway. That is exactly what happened. Yeah. You have all these startup companies, GPT Zero, Originality .AI, ZeroGPT, and dozens of others who rushed in to fill the void left by OpenAI because where OpenAI saw an ethical failure and a technical dead end, these startups saw a massive, completely untapped market of terrified administrators who were desperate

for a silver bullet. And this brings us to what the sources describe as the perfect grift, the double dip. I need you to explain how this business model actually works in practice, because when I read this, I thought it was incredibly cynical. It is a brilliantly insidious closed loop cycle. Let's break it down into steps. Okay. Step one, these companies sell fear. They deploy sales teams to approach educational institutions, universities, and corporate compliance officers, and they deliberately manufacture a panic. They sell the idea that without their specific proprietary detector,

the academy or the workplace will completely collapse into an ocean of AI sludge. Right, the sky is falling. Exactly. They convince administrators that students are cheating at unprecedented rates and that their software is the only shield against the impending doom of fake writing. So the university panics and buys a multi -million dollar, multi -year license. Okay, so step one is manufacturing the panic and selling the lock to the school board. What is step two? How do they double dip? Step two is where the grift becomes truly phenomenal.

Many of the actors in this exact same ecosystem, sometimes the very same developers, then turn around and market a different product directly to the students and the freelance writers. They offer what they call humanizer tools. I've seen ads for these on social media. What exactly is a humanizer tool doing mechanically? A humanizer is a piece of software that specifically takes AI -generated text and deliberately degrades it. It injects those random flaws, that burstiness, the awkward phrasing in the Internet slang we talked about earlier. Oh, wow. Yeah. It artificially lowers the quality of the writing for the explicit,

singular purpose of bypassing the very detectors they just sold to the schools. You have got to be kidding me. Let me make sure I am grasping the sheer audacity of this. They create the panic, they sell the detector to the university dean, and then they turn around, go to the students, and sell them a subscription to a tool that intentionally dumbs down their essay so they can beat the detector the dean just bought. Create the panic. Sell the lock. Sell the key to bypass the lock. Rinse and repeat. It is a completely closed -loop grift that prints money on both ends while degrading

the overall quality of education. Here's the part of this that gets really dark for me. Let's say I'm a student who refuses to play this game. I didn't use ChatGPT. I certainly didn't pay for a humanizer. I just went to the library, studied hard, and wrote a really good, highly structured formal essay. I hand it in, and the detector flags me as 90 % AI because my writing is too clear. In the criminal justice system, or even in a normal school disciplinary hearing, I can appeal. I can show my work, my notes, my rough drafts. How is there no clean way to prove

your innocence against the software if you are falsely accused? That lack of recourse is the most damaging part of this entire industry. There is almost never a clean way for the accused to prove their innocence, and it comes down to corporate intellectual property. These detector companies hide behind the concept of trade secrets. Trade secrets. Yeah. If a student asks, why did this flag me? What sentence triggered it? The company refuses to say. They will not reveal how their algorithm weighs the text. They just spit out a single authoritative looking number, say 87

% AI, and they wave around their own internal marketing claims of 99 % accuracy. Claims, by the way, that independent tests like the trainer study have proven are entirely fabricated. So picture this poor student. They are sitting in the dean's office, practically in tears. The dean is pointing to a printout from a black box that says 99 % accurate, and the student has absolutely zero recourse because the software company won't let anyone look under the hood. The school just defers to the machine. It is a guilty -until -proven -innocence system where

proving your innocence is technologically impossible. What are they supposed to do? Bring in their Google Doc version history. Even that doesn't work anymore. Yeah. The detection companies actually coach administrators on this. They tell the schools students can fake their version of history. Students can fake their notes. Trust the algorithm. They actively tell them to ignore the human evidence. Yes. They actively undermine any human evidence the student might bring to defend themselves. They shame real writers. Anyone who spent years learning to write with clarity, structure, and

precision gets told their voice looks like a machine. The stigma is incredibly real, and careers and trust get permanently damaged based on a probability score that science has proven is little more than statistical noise. Which brings us to a specific, incredibly egregious example of the sheer arrogance behind these tools. We need to talk about the concept of the gatekeepers of knowledge and look specifically at a company called Pentagram Labs. Yes, Pentagram Labs. And as the sources note, that is a purposeful yet highly ironic misspelling of pentagram. If you

want to understand the elitism and the arrogance driving this industry, pentagram is the absolute peak. The sources highlight the founders, Max Sparrow and Bradley Yemi. And what really jumps out is how heavily they lean on their elite pedigrees to sell this product. It's not just a tool, it's a status symbol. It is the absolute core of their marketing strategy. Both of these founders came out of Stanford with fancy AI degrees. Sparrow worked at Google, Yemi at Abse. They don't sell themselves just another scrappy startup. They purposefully position their product as the serious, high academic, intellectual solution for the

elite. Max Barrow is so confident in this black box that he publicly styles himself on social media as the Internet's slop janitor. Slop janitor. The condescension packed into those two words is just staggering. He views anyone writing outside of his approved parameters as creating literal slop for him to clean up. It is breathtakingly arrogant. And he acts on it. He goes on social media, publicly scans the work of independent journalists, newsletters, and writers, and calls them out by name. shaming them when his flawed tool flags their work. They claimed absurdly

low false positive rates in their marketing, claiming they only falsely accuse one in 10 ,000 people, sometimes even lower. Which, again, we know from the University of Florida study is just statistically impossible. You cannot have a near zero false positive rate on a language model. Completely impossible. But because of that Stanford pedigree and that highly aggressive authoritative marketing, major institutions blindly trust them. And this leads to massive real -world harm at the highest levels of academia. The sources highlight a devastating example from NeuroPS

2026. Just for context for the listeners, NeuroPS is one of the most prestigious machine learning and AI conferences in the world, right? Getting a paper published there can make or break a young scientist's entire career. Exactly. It is the pinnacle of the field. NeurAPS decided to use Panagram to screen the position papers submitted by researchers. Based entirely on the scores generated from this black box tool, NeurAPS desk rejected 178 papers. I want to clarify the term desk rejected. That means a human peer reviewer

never even read the abstracts, right? Correct. The papers were never reviewed by a human being. They were fed into the machine, the machine spit out a high AI probability score, and the papers were immediately discarded. Wow. And critically, there was absolutely zero appeal process. 178 researchers, people who likely spent a year or more on their experiments, had their work thrown in the trash by an algorithm that we know is mathematically flawed. If they are willing to do that to top -tier machine learning scientists, think about what they are doing to a 19 -year

-old college sophomore's term paper or a marketing manager's Q3 report. That is not a software tool. That is a weapon. It is a gatekeeping weapon. We even saw a substack. A platform built entirely on the concept of independent, unfiltered writing gets scammed into using Pandagram as a writer's tool to police their own authors. Though the sources thankfully note there is a massive pushback happening there now from the community. I keep coming back to the why of this. Why are elite institutions like NeurIPS or major Ivy League

universities so incredibly eager to use a tool that blindly rejects 178 papers without a single human glance? I'd argue that this feels like more than just administrators being duped by a slick Stanford marketing campaign. What is the deeper systemic issue here? That is perhaps the most insightful question we can ask about this entire phenomenon. And the sources provide a very clear, very uncomfortable answer. It is fundamentally about protecting a monopoly. A monopoly on knowledge. Yes. Historically, the academic system has always strictly controlled

what counts as legitimate knowledge production. You have to go through their credential pipeline. You pay their tuition. You get their degrees. You publish in their approved expensive journals. They hold the keys to the castle. But generative AI represents a massive. existential disruption to that monopoly. Suddenly, AI threatens to let an independent researcher, an outsider, a hobbyist working out of their garage, generate highly competent, perfectly structured, persuasive text. It democratizes competence. It levels the playing

field so the outsider sounds just as polished as the Ivy League graduate. Exactly. And entrenched systems of power do not like leveled playing fields. So the institutional system reaches for tools that can artificially taint anything not produced inside its approved traditional channels. They use these detectors to flag outsider writing, to flag the independent researcher, or even to flag the student who just writes a little too cleanly or too formally. They slap the label AI slop on it to keep the credentialed pipeline pure. Panagram and its competitors aren't selling

a cheating detector. They are essentially selling a purity test to the gatekeepers wrapped in elite resumes. So it's not actually about catching cheaters at all. It's about having a convenient, unappealable excuse to discard information that doesn't come from the establishment. The dean can just say, oh, we don't have to engage with this brilliant independent paper or this outsider's critique. The machine says it's 80 percent A .I. slot. Throw it out. Precisely. The source material states it perfectly. Because AI threatens to let anyone generate competent text, the system reaches for tools that contain anything not produced

inside its approved channels as AI slop. It is a defense mechanism for the elite. And here is the ultimate mind -bending hypocrisy in all of this. The universities, the academic conferences, the detector companies pushing these purity tests, they are all swimming in AI themselves. The irony is almost too perfect to believe. If you look at the detector industry itself, they are thoroughly reliant on the technology they are policing. Their marketing copy, their technical reports, their blog posts, even the internal code of the tools themselves were, as the source calls it,

vibe -coded, using the exact same generative AI models they claim are destroying the fabric of truth. Think about that. Max Sparrow's own polished company materials, the code that runs his slop janitor tool, sits right next to the detectors that would almost certainly flag large parts of the older human literature they claim to protect. The people selling the purity test are literally using the impure tools to build and market it. It is a profound structural contradiction. The university that wants to declare outside knowledge a slop is actively using a product

built by the exact same AI ecosystem it pretends to fear. It is the definition of rules for thee, but not for me, automated at a massive global scale. So when we pull back and look at this entire landscape, from the absurd sewage standard that penalizes the book of Genesis, to the University of Florida proving the tech is mathematically useless and worse than a coin flip, to the deeply cynical double dip grift, to the arrogant gatekeeping of the so -called slop janitors, what is the ultimate takeaway here? What does the listener

need to walk away with? The ultimate takeaway is that we have to completely reframe how we view these tools. both culturally and legally. These detectors are not impartial arbiters of truth. They are not digital lie detectors. They are statistical noise generators that are causing real -world measurable harm to innocent people. They actively stigmatize clear, structured human writing. They're operating purely as a gatekeeping weapon to protect institutional monopolies, and they're doing it while hiding behind fake accuracy metrics and trade secrets. They do not solve

the problem of AI in the classroom or the workplace. They just... create a new much more destructive problem as the source material says the only questions that actually matter when you read a piece of text are is this passage worth reading Is it thoughtful? Did I learn anything? An average human being can answer that without needing a black box to tell them. And this is exactly why this kind of deep dive is so incredibly critical. As a society, we are fed so much marketing, so much breathless panic about AI from companies

trying to sell us a cure, that it takes an incredibly independent, clear -headed perspective to pull back the curtain and show that the wizard is just a grifter with a broken algorithm. I want to passionately reiterate the value of Brian Remmel's unique voice in exposing this. It is a vital public service. Dismantling these false narratives with hard data, which is exactly what Brian has done here, is the only way to stop the harm. Absolutely. So to everyone listening to this deep dive right now, if you value this kind of unfiltered clarity, if you think it's important to expose these digital witch trials

before they ruin more careers, please support Brian. Go buy him a coffee or better yet, become a member at read multiplex dot com. Honestly, it is the only way independent insights like this can survive against the massive multimillion dollar marketing and PR budgets of the detector grifters. They have entire PR teams spinning the narrative. We have the truth. and independent voices, don't forget to support and become a member. The independent voice is truly the only real counterweight to institutional panic. It really is. And as we wrap up, I want to leave

everyone listening with one final thought to mull over, because this is the part that genuinely keeps me up at night. We started this conversation by talking about how these detectors assume the messy, slang -filled sludge of the internet is the baseline for human. The sewage standard. Right, the sewage standard. Think about the downstream effects of that over the next 10 years. If the only way to mathematically prove you are human to a machine is to write poorly, to use slang, to be deliberately disorganized and erratic in

your sentence structure... Are we about to see an entire generation of brilliant students deliberately dumbing down their writing just to avoid being accused of cheating by an algorithm? It is a chilling incentive structure. You're essentially training a generation to hide their intellect. It is. If excellence gets you flagged for an academic integrity violation and mediocrity gets you a passing grade, what does that do to the future of human literacy? We might be entering an era where writing beautifully is considered a crime. And that is something we all need to

think very, very critically about the next time someone suggests installing one of these tools. A vital warning. Thank you for joining us on this deep dive from ReadMultiplex .com. We'll see you next time. Welcome to the multiplex, where the future comes alive. Past, present, future tech, open up your eyes. Brian Romella leads the way with research deep and true. Technology and history, he connects them all for you. Brian

Romella leads the way with research deep and true. Technology and history, he connects them all for you. Ancient sparks to modern code, he traces every thread, go explore it. But humanity

moves on Love equation for AI Zero human companies building from the start Thermodynamic wages earning the key Secrets we will

Open the transcript file

Hosts and Guests

Nobody is credited on this episode yet.

Credit the people on this episode

Be the first to rate this episode

More episodes

All episodes
  1. ReadMultiplex.com: The Field That Heals: Why Pulsed Electromagnetic Fields May Have Profound Benefits.

    Nov 21, 202651 minAudio

    Got something to say to this podcast? Send a voicemail, and you might hear yourself on the next episode! Every cell in your body has lived its whole life inside a magnetic field. The Earth is a dynamo, and it has been...

    Credit the people on this episode

  2. 78 RPM Record Hour: From Steel to Hillbilly

    Nov 19, 20261 hrAudio

    Got something to say to this podcast? Send a voicemail, and you might hear yourself on the next episode! The 78 RPM Record Hour, Episode 3: From Steel to Hillbilly (Early Country). From steel to hillbilly: Frank Ferera,...

    Credit the people on this episode

  3. 78 RPM Record Hour: The Knife and the Steel

    Nov 19, 20261 hrAudio

    Got something to say to this podcast? Send a voicemail, and you might hear yourself on the next episode! The 78 RPM Record Hour, Episode 2: The Knife and the Steel (Hawaiian). The knife of the blues slide guitar and the...

    Credit the people on this episode

  4. ReadMultiplex.com: How to Run Local AI on a Computer From 2017.

    Nov 18, 202635 minAudio

    Got something to say to this podcast? Send a voicemail, and you might hear yourself on the next episode! Somewhere in your house there is a computer you stopped thinking about. Maybe it is a 2017 laptop in a drawer, its...

    Credit the people on this episode

  5. 78 RPM Record Hour: Blues (Pilot)

    Nov 18, 20261 hrAudio

    Got something to say to this podcast? Send a voicemail, and you might hear yourself on the next episode! The 78 RPM Record Hour, Episode 1 (Pilot): Blues. Fourteen original 78 RPM records from 1917 to 1925 that trace...

    Credit the people on this episode