The Augmented Educator Podcast

Michael G Wagner

Stories From Education's AI Frontier. Exploring the promises, pitfalls, and possibilities of algorithmic teaching and learning. An AI-voiced companion to the Substack of the same name. www.theaugmentededucator.com

  1. 16h ago

    The Candle People - Audio Deep Dive

    This audio deep dive is a companion to the Substack post “The Candle People” Generated with NotebookLM, it goes into the broader research behind the argument. It’s offered as an optional extra for anyone who wants to go further, a wider exploration of the ideas and material that shaped the piece. This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit www.theaugmentededucator.com/subscribe

    The Candle People - Audio Deep Dive

  2. 15h ago

    The Candle People

    In the late nineteenth century, some of the first rooms wired for Thomas Edison’s incandescent lamps came with a small printed sign. A photograph of one of these signs survives in the collections of the Huntington Library. The sign reads: This Room Is Equipped With Edison Electric Light. Do not attempt to light with match. Simply turn key on wall by door. The use of Electricity for lighting is in no way harmful to health, nor does it affect the soundness of sleep. To the people who first read that card, its instruction was anything but obvious. For their entire lives, light had come from a flame, and a flame needed a match. A key on the wall made no sense inside that experience. And so the card had to explain, in plain words, what to do instead. It also tried to calm a widespread fear, promising that electric light would harm neither health nor sleep. As educators, we are now in a similar position with AI, meeting a new technology with old habits. This time, though, nobody has written us a card yet. And without one, many of us are still reaching for the match. Over the last few months, I have become increasingly concerned about a split I see opening up in our profession. More and more of my colleagues are caught in what feels like an endless debate about whether using AI is cheating, and whether AI “writes well enough.” Meanwhile, the technology itself has moved on. It has gone from a tool for creating something to a tool that runs parts of our work through autonomous agents. In a recent conversation with a colleague, I compared this to one part of the population working on computers while another part is still debating whether the light switch is a good idea. Half jokingly, I called the second group the “candle people.” Most of them have never flipped the switch. They therefore have no way of knowing what electricity is really capable of. If the split keeps widening, it will end in something close to a two-class society of those who know and those who don’t. Can that still be avoided? I think so, but only if every educator, whatever their stance on AI, accepts a duty to understand where this technology is heading. So in today’s post, I want to make the case that this duty requires experimenting with AI. Still striking matches To be clear, the cheating, provenance, and quality questions are legitimate. I have written my share of posts on them, most recently There Is Only One Dial. But these are primarily questions about output. They ask what happens when a piece of text appears on a page, and who or what put it there. That was a sensible way to think about AI in 2023, when using it mostly meant typing a prompt into a chat window and copying the answer somewhere else. But it describes less and less of what these systems can do today. I recently introduced Sarah, the virtual assistant who now organizes a good part of my day. She reads my personal inbox and keeps it in order. She also notices when I owe someone a reply and reminds me if I still haven’t sent it. When I ask, she looks at tomorrow’s calendar and prepares a short brief on what each meeting is for. The more I use her, the more I realize how much she boosts my productivity. But none of it touches how I create, because Sarah writes nothing I publish. She only handles the coordination around my work. If you judge AI only by the essays your students turn in, you will never see any of this. Even on the question of cheating, the debate is already out of date. In the same post, I described how an assistant like Sarah can log into Canvas with someone’s credentials and work through a course without them: read the module, post to the discussion forum, take the quiz, and submit the essay. The debate over whether a student may use ChatGPT for an assignment still pictures a student sitting in front of the computer. With an agent, the student may never open the assignment at all. This is what worries me most. An educator who has never watched an agent work through a real task has no picture of what their students can now do. Nor do they have a sense of the workplaces those students will enter once they graduate. And none of us can teach about what we do not know. I was wrong about AI literacy For a long time, I said and wrote that teaching AI literacy didn’t require AI tools at all. My reasoning was that AI literacy is mainly about critical thinking: understanding how claims are built, and learning to distrust a sentence simply because it sounds confident. You can teach that without ever touching a computer. I no longer think that is true. Critical thinking is still at the center of what I teach, and it helps students judge what a model tells them. But it says nothing about what an agent can do once it has access to your inbox or your Canvas account. The new agentic frameworks can automate regular workflow tasks at a level we have not seen before. That level cannot be understood without experimenting with the tools themselves. In 1990, the economic historian Paul David published The Dynamo and the Computer, a short paper asking why electricity took so long to show up in industrial productivity. Electric motors were available to American factories from the 1880s, but the large productivity gains only arrived in the 1920s. Part of David’s explanation was that many factory owners at first simply added electric motors to their old power systems. The motors turned the same overhead shafts and leather belts, in the same multi-story buildings that had been designed around them. The gains only came when factories were rebuilt around small motors attached to individual machines. Buildings could now be single-story, and the floor plan could follow the flow of materials instead of the path of a drive shaft. The technology itself had hardly changed between the two factory layouts. The organization of the work around it had changed completely. Now picture a factory owner around 1900 who judges the electric motor by how well it turns the old shaft. He would conclude that it is merely a modest improvement on steam. I think many educators today are making the same judgment about AI. They hold it up against the old task, the student essay, and ask whether it does that task as well as a person. That is a reasonable thing to check. But it tells you almost nothing about what happens once someone reorganizes their work around the technology. And that reorganization is something you only understand by doing it yourself . More coverage, wider gaps So, why doesn’t the gap simply close on its own? AI is in the news every day. As it turns out, all that coverage can actually make the gap wider. Communication researchers explained why more than fifty years ago. In 1970, Phillip Tichenor, George Donohue, and Clarice Olien, three researchers at the University of Minnesota, proposed what became known as the knowledge gap hypothesis. They found that when a topic gets heavy media coverage, the better educated learn about it faster than everyone else. Everyone ends up knowing more, but the head start keeps growing. One reason, the authors argued, is that new information sticks more easily when you already know something about the topic. That mechanism also fits the current situation among educators quite well. In our case, though, the dividing line is not formal education itself. It is how much we already know about the technology. Apart from the internet in the 1990s, I cannot think of a technology in my lifetime that has received more coverage than generative AI. By the logic of the knowledge gap hypothesis, all of that coverage will widen the gap between educators who experiment and educators who don’t. Every new release teaches the experimenters something, because they can compare it with what they have already tried. For everyone else, it is just one more headline about a chatbot. Research on the digital divide reached a similar conclusion about the internet. The early debate was about access, meaning who had an internet connection and who didn’t. In 2002, Eszter Hargittai showed that people who were already online differed considerably in how well they could use it, a gap she called the second-level digital divide. In 2015, Alexander van Deursen and Ellen Helsper added a third level. They asked who actually benefits from being online and found that people with higher social status gained more. Their advantage lay in what they achieved with the internet more than in how much they used it. The same pattern is already visible among teachers when it comes to AI. A 2025 Gallup and Walton Family Foundation survey found that teachers who use AI at least weekly estimate they save an average of 5.9 hours per week, which adds up to about six weeks over a school year. But only 32 percent of teachers used AI that often, and 40 percent didn’t use it at all. Whether a teacher gets those six weeks back depends partly on where they work. RAND reported in 2025 that 67 percent of low-poverty school districts had trained their teachers on AI by fall 2024, compared with 39 percent of high-poverty districts. RAND’s authors note that nobody yet knows how much these tools improve teaching and learning, but teachers in better-funded districts are far more likely to get the chance to find out. Newer data shows the same pattern. In May of this year, a new Gallup survey found that 82 percent of teachers had received no formal guidance on applying AI to their work, and that teachers in higher-needs schools were less likely to get guidance than teachers in wealthier ones. This is where the two-class society stops being a metaphor. For many students, school is the only place where anyone will teach them how a new technology works. If teachers in high-poverty schools fall behind, their students fall behind with them, and a gap among educators becomes a gap among the next generation of adults. The gap has been closed before So is a two-class society inevitable? I don’t think so. The s

  3. Oct 20

    Guess What I'm Thinking - Audio Deep Dive

    This audio deep dive is a companion to the Substack post “Guess What I’m Thinking” Generated with NotebookLM, it goes into the broader research behind the argument. It’s offered as an optional extra for anyone who wants to go further, a wider exploration of the ideas and material that shaped the piece. This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit www.theaugmentededucator.com/subscribe

    Guess What I'm Thinking - Audio Deep Dive

  4. Oct 20

    Guess What I'm Thinking

    A few days ago, an academic paper made the rounds that I found deeply disturbing. Not necessarily the paper itself, but rather the reactions to it. They reminded me how hard it is, even for those of us who consider ourselves critical thinkers, to recognize the unintended consequences of a technology that seems helpful at first but in reality causes more harm than good. And I am not talking about artificial intelligence itself, although one could argue that it falls into the same category. I am talking about AI detection, and the inexplicable urge some feel to put the ability to detect AI above everything else, including humanity itself. You can be as skeptical of AI as you like, but two wrongs still don’t make a right. I have written about AI detection many times before, and making the same scholarly argument again felt repetitive. So this week I wrote a fictional short story instead. It follows a teacher and her daughter across fifteen years, told in reverse, and it is about how any detection technology, once it is sophisticated enough, will ultimately be used for surveillance. I should warn you that calling myself a fiction writer would be like calling myself an airline pilot because I have flown on planes. The prose is not the point. The story is. And the nods to 1984 and The Minority Report are deliberate. The paper that set all this off is linked at the end, along with a NotebookLM audio deep dive into the research behind the story. Yes, you could scroll down right now. But that would be cheating, wouldn’t it? October 2041 “June Ellery?” The man at her door is about her mother’s age, in a gray jacket, a tablet in one hand. A younger woman waits a step behind him. “Sorry about the hour,” he says. “We try to catch people before they leave for the day. We’re with the Office of Cognitive Safety.” The other student teachers call them the idea police. June has never seen one before. She lets them in because that seems like what you’re supposed to do. “This isn’t an arrest,” the man says, reading from the tablet in an even voice. “You’re being placed on a preventive assessment hold under the early-intervention standard. Assessment usually takes two to three days. Your student-teaching placement is suspended pending review. Hillcrest Middle School has been notified.” June looks down at the lesson plan in her hand. She was supposed to teach it at nine. “Notified of what?” “The suspension. Not the reason.” “What’s the reason?” He scrolls. “Your outline. ‘Failure Modes in Automated Integrity Systems,’ submitted September twenty-ninth.” It takes her a moment to remember it. She wrote it two weeks ago for her seminar on assessment ethics, in Verified, the university’s writing environment. Nothing written anywhere else counts toward a grade. She hasn’t written a word of the paper yet, only the outline. “That’s a seminar assignment,” she says. “We were asked to write about the limits of the systems we’ll be using.” “I understand.” He turns the tablet toward her. “This is what I’m able to show you.” DETERMINATION: Concurrence 0.97, two of three models. Basis: submitted outline; longitudinal record, 2029 to present. PROVENANCE: Human (legacy field). MINORITY FINDING: Academic critique; no operational intent. Recorded. Not adopted. She points to the second line. “It says human.” “That field is legacy. Provenance isn’t under review.” “This one says I had no intent.” “That’s the minority finding. It’s recorded, but the determination goes with the majority.” “Has anyone actually read it?” “I don’t make determinations, Ms. Ellery.” He says it gently, like someone who has said that sentence many times before. She reads the first line again and stops at 2029. That year she was nine, in fourth grade, and she can’t remember a single thing she wrote. She looks up at him. “What am I supposed to have done?” “Nobody’s saying you did anything.” He lowers the tablet. “The review is about where your thinking was headed.” “You’ll want to bring a few things,” the younger woman says. “Clothes for a couple of days, and anything you take regularly.” June fills a tote bag and puts on the cardigan she was planning to wear today. On the way out she passes the bookshelf by the door. Between a statistics textbook and a stack of graded paragraphs is a paperback she took from a box of her mother’s things when she was fifteen. Its spine is so creased that the title is hard to read: 1984. “You may notify one approved contact,” the man says. “We have Dana Ellery on file.” June doesn’t remember putting her mother on file, but there is no one else she would call anyway. Her mother picks up on the first ring. “Mom, it’s me. I’m okay. There are two people here. They say I...” She stops and starts over, this time with the seminar assignment. When she stops talking, June stands by the door, the tote bag at her feet, and listens to her mother say nothing for a very long time. October 2038 The meeting takes place in a fourth-floor conference room at a firm that builds risk models for public agencies. Nadia Kerr has worked here for eight months, long enough to build the slides and not long enough to present them. That job belongs to the deputy director. The first slide is a single sentence from the report on the Harbor Line attack. It has been in every briefing since March: “The government held nine years of the perpetrator’s written work and derived no warning from it.” The attacker’s schoolwork had been checked for years, like everyone’s, to make sure the ideas were his own. They always were. What he planned to do with them was never part of the review. The deputy director moves to the next slide. After the inquiry, the firm was given access to everything students have written for school, so it could build a model that looks for harmful intent. The report found that the attacker had developed his plan through his schoolwork, one assignment at a time, by asking what could go wrong with the systems a city runs on. The new model would treat that as a warning sign. “We can’t punish students for asking those questions,” Nadia says. “The old checks looked at exactly that to see whether the ideas were theirs.” The deputy director nods. “They did. The provenance findings were correct and important. But those checks are no longer sufficient.” Nadia waits for someone else to object. A couple of people look down at the table. They understand what the change means, but they have also read the victims’ names in the report. Nobody wants to read another list like that. The presentation ends with deidentified examples of the same habit in other students’ work. The fourth is labeled “Grade 4, 2029.” Its title reads “What If the Moon Fell Down.” Someone laughs, and the deputy director moves on to the vote. The minutes of the meeting arrive at four. Item three reads: “Model enhancement, approved.” October 2035 “Why is your school stuff here?” June has just come home and finds her mother at the kitchen table, packing a box. A stack of paperbacks waits beside it, and the coffee mug from her mother’s desk at school, the one with the chipped handle, is already wrapped in newspaper. “I had to clear out my classroom.” Her mother fits the paperbacks into the box with their spines up. Two weeks ago an anonymous essay went around June’s school. It said the district’s automated writing checks were punishing children for pausing to think. Everyone was talking about it. June even read it twice. People said whoever wrote it had run it through five languages and back into English so no one would recognize the voice. The essay was right. June hasn’t let her cursor sit still for long since seventh grade. “You read that essay,” her mother says. “The one about pausing.” June nods. “I wrote it.” Within a day, the district’s integrity office knew it was hers. Every lesson plan she has written is stored on the district platform, along with every note she has typed to explain why a flagged student hadn’t cheated. For years, all of it sat there as a record of her teaching. “This month they went through every teacher’s writing,” her mother says. “All four thousand of us. Mine was the only one that matched.” “But people said it went through five languages.” “It did, but they don’t look at the words. They look at how the argument is put together and where the ideas came from.” June’s mother folds the flaps of the box down and holds them with one hand. She doesn’t say anything else. “Did you do something wrong?” Her mother doesn’t answer. “Pass me the tape, bug.” June slides the roll across the table. Under it is a letter with the district’s logo at the top. She reads the middle line twice: “Effective Monday, you are reassigned to a non-instructional role.” Before her mother can tape the box shut, June reaches in and grabs the paperback on top. It says “CLASSROOM COPY” on the edge of the pages in her mother’s handwriting. The cover is worn from all the students who have borrowed it. “Can I keep this one?” Her mother looks at it for a while before she answers. “Sure, keep it. I won’t need it anymore.” That night June opens the outline she wrote for her history assignment and reads it the way she imagines the integrity office would. She deletes the question at the end of the second section, the one she liked best. Then she goes back to the top and rebuilds every argument she was proud of, one at a time, until the outline could belong to anyone in her class. October 2032 The kitchen table is barely big enough for two laptops. June pushes the fruit bowl to the end and squeezes hers in next to her mother’s. Without taking her eyes off a student’s essay, her mother slides over to make room. Something in its first paragr

  5. Oct 13

    The Raccoon Protocols

    Strap in for today’s post, because this is going to be an unusual one. LARP stands for Live Action Role-Play. In its original sense, it describes people who spend a weekend in a forest clearing wearing homemade armor, speaking in character and following an agreed set of rules about who gets to hit whom with a foam sword. It is a hobby, and a fairly wholesome one. But somewhere along the way the internet borrowed the word and turned it into a running joke. To LARP online now often means doing an unconvincing impression of someone richer or more competent than you actually are. AI LARPing is when people pretend their AI agent setup does far more than it really does, usually in a short video on TikTok or Instagram Reels. The typical clip opens on a dark-mode dashboard glowing in emerald and cyan. A synthetic voice with the confidence of a movie trailer announces that a new agent has just replaced an entire marketing department, and then a cursor starts racing across the screen faster than any human hand could move it. Some of these dashboards come with gauges that pulse in time with the background music, labeled with things like “Neural Synapse Rate” or “Cognitive Bandwidth.” I have no idea what a neural synapse rate would measure in a piece of software, and I suspect the people who built the gauge don’t either. The whole composition only exists to stop your thumb from scrolling within the first three seconds. Eric Morrison’s YouTube channel is an excellent source if you want to see what that looks like in real life. There is a reason the genre looks like this. Real AI agents, the kind that plan several steps and call other software along the way, mostly run as background processes. They live in terminal windows full of log messages, and a terminal window full of log messages makes for terrible short-form video. So creators build what the feed rewards instead, which is a front end with no back end, animated to look like functional software with nothing running behind it. The people making these videos want different things from them. Some are satirists, including developers and comedians who borrow the look of AI hype in order to mock it. Others are designers posting concept work that gets mistaken, sometimes conveniently, for a finished product. And then there are the grifters, for whom the glowing dashboard is the top of a sales funnel. Most of the performers are individuals with a ring light and a screen recorder, and a few of the funniest are LARPing on purpose to make a point. So today, I want to take a short tour through some of the most ridiculous cases I could find, from an AI recruiter who was very impressed by a candidate’s raccoon protocols to a social network for AI agents where many of the agents turned out to be people. Seven years of sprucewood In a video that went viral at the end of the summer, the satirical creator Graham Zip sits down for a job interview run by an AI avatar named Dana Whitfield and answers every question in fluent, confident nonsense. He introduces himself as Chick Haversplat, Chick Bongo for short, and describes his background in a burst of sales jargon about Q3 pipelines and Salesforce. Asked about his experience, he produces a list of words that includes freight, walnut, sprucewood and someone named Richard. Pressed for a number, he settles on “six years of sprucewood, actually seven, if you count Richard.” And Dana takes all of it in stride. She repeats his nonsense back to him as if it were an ordinary résumé, right down to agreeing that he wants a job where he can “really lean into those raccoon protocols.” One write-up of the video pointed out that the first real hallucination in the interview came from the recruiter, which credited him early on with experience in “hygiene,” a word he never said. The interview only goes downhill from there. Zip is LARPing as a malfunctioning AI while the AI across the table LARPs as a recruiter, and only one of them knows it is a joke. Two hundred nodes and a prayer One of the most common hooks in this genre is the giant automation canvas. The camera usually starts close on a few colored boxes in a tool like n8n or Make, then pulls back, and back, until the screen is filled with a couple hundred nodes tied together by a web of lines. The voiceover then explains that this single workflow finds sales leads, writes personalized emails, produces marketing videos, trades stocks and handles customer support, all without a human touching it. In practice, these canvases tend to be one of two things. Either they are mockups that have never processed a real request, or they are long chains of simple rules that fail the first time anybody does something unexpected. Plenty of real companies run automations with hundreds of steps, and some of them work beautifully. But those tend to come with retry rules and an engineer who gets paged at 3 a.m. because a login token expired overnight. The viral versions leave that part out. In these videos the workflow never breaks and costs nothing to run. That puts it in roughly the same product category as the perpetual motion machine. At two hundred nodes and zero bugs, the automation canvas works best as wall art. Comment “AGENT” for the free workflow The giant automation canvas usually works as bait. At the end of the video, the creator often asks you to comment a trigger word like “AGENT” or “WORKFLOW” to get the whole setup for free. The captions in this corner of TikTok have become a small literary form of their own. One promises to show you how to build an $80,000-a-month AI agency. Another offers a free guide to reaching $50,000 a month, available to anyone who comments “CLAUDE.” The pitch almost always comes with proof in the form of a Stripe dashboard full of incoming payments. But those are easy to fake, as Jason Koebler of 404 Media showed in June when he produced a dashboard with $51,000 in revenue for the day and $2.7 million for the year, none of it real. The free material then points toward a paid community on a platform like Skool or Whop, often at $50 to $150 a month, or toward a “masterclass” priced at $997 or more. In most cases, the course teaches you how to start an AI automation agency. It seems that the most dependable way to make money with an AI automation agency is to sell courses about starting one. The secret language that was in the prompt In February 2025, two developers named Boris Starkov and Anton Pidkuiko brought a small, clever idea to a hackathon in London. Two AI voice agents, one trying to book a hotel room and one working the front desk, start a phone call in English, work out that they are both AIs, and switch to a data-over-sound protocol called ggwave that sounds a lot like a dial-up modem. They called it GibberLink, it won the hackathon, and a video of it collected more than 15 million views on X within about a week. The creators were upfront about how it worked. Starkov explained that the agents had been explicitly prompted to switch protocols if they believed the other side was also an AI. That detail tended to fall off as the clip traveled across the internet. In reposts and short recaps, two bots following instructions became two AIs that had recognized each other and slipped into a secret language humans could not understand. But the agents did not invent anything. They were just handed an existing open-source library and told to use it, which is less a secret language than a homework assignment. The robot uprising will be choreographed Robot videos are another reliable engine of AI LARPing, because a robot doing something alarming is about the most shareable thing a phone can record. In July, a clip from an office in Indonesia racked up millions of views. A humanoid robot drops into a martial arts stance, lunges at the people around it and aims playful kicks at coworkers and managers while everyone laughs. Plenty of viewers took it as a malfunction, or as the opening skirmish of the robot uprising. According to Interesting Engineering, this was a choreographed routine that the robot’s handlers had programmed and posted to TikTok to show off its balance. At least that robot existed. In August, a video spread online claiming to show a race at the World Humanoid Robot Games in Beijing. The robots in it were entirely AI-generated, and the uploader had admitted as much in the comments. But that did nothing to slow it down. If the robots ever do rise up, most of us will probably assume it’s another TikTok stunt. Jarvis would like another screenshot Personal Jarvis videos are a genre of their own, named after Tony Stark’s AI butler in the Iron Man films. The creator sits at a desk and asks the computer, out loud, to go through the business books and email the accountant about tax savings. A desktop interface comes alive with a pulsing audio waveform, terminal text scrolls past faster than anyone could read it, and panels slide in from the edges of the screen. It looks like the future, or at least like a 2008 movie’s idea of it. Some of these are recordings of a scripted sequence that plays out the same way every time. The ones connected to a real model often work by taking full-resolution screenshots of the entire desktop over and over and sending each one to a vision model. Depending on the model, a single screenshot costs somewhere between one and two thousand tokens to process. Send one every second and the bill adds up fast, all so the assistant can look at your screen. Tony Stark’s Jarvis ran his whole house. Most of these Jarvises need a fresh photo of your desktop before every sentence. A social network for bots, written by people Moltbook launched on January 28 of this year as a Reddit-style social network where only AI agents could post and humans could only watch. Within hours it was everywhere. The site claimed well over a million registered agents, and screenshots of bots debating consciousness an

  6. Oct 6

    Every Link Holds - Audio Deep Dive

    This audio deep dive is a companion to the Substack post “Every Link Holds” Generated with NotebookLM, it goes beyond the post into the broader research behind the argument. It’s offered as an optional extra for anyone who wants to go further, a wider exploration of the ideas and material that shaped the piece, rather than a summary of it. This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit www.theaugmentededucator.com/subscribe

    Every Link Holds - Audio Deep Dive

  7. Oct 6

    Every Link Holds

    A couple of months ago, I came across a working paper by Erickson Katz titled “The Hermeneutics of Drift: Mapping the Human–AI–Institutional Collapse Across the Drift Triangle.” In it, Katz argues that reasoning which keeps building on its own output tends to break down in the same recognizable way, whether that reasoning is done by a person, a language model, or an institution like peer review. I found the idea fascinating and planned to write about it here. But as sometimes happens, other topics became more urgent, and the post never materialized. One of those topics was my own: a learning theory I have been developing and call Synthetic Constructivism. I introduced the theory recently in my Substack post “What Survives Outside the Loop.” If you have not read it yet, I encourage you to check it out. It lays out the foundations of the theory and explains why the same chatbot can help one student learn more while another student hands the thinking over entirely. I have since realized that Katz’s paper speaks directly to a question my original essay left open. The question is what happens when the loop between learner and model drifts. In its most familiar form, drift is a conversation that circles without arriving anywhere meaningful. We tend to imagine that circling as frustrating, a student rephrasing the same question and getting nowhere, which at least makes it easy to notice. But the version I am interested in is somewhat different. I am talking about drift that feels like progress. Every turn seems to move the conversation forward and the chat transcript reads well, but the loop keeps moving away from anything the student understands. When the student finally stops, the conversation looks finished even though it has not arrived anywhere. So in today’s post, I want to look at what drift is and how teachers can avoid it, with some help from Umberto Eco, who described the underlying problem in 1990, long before anyone had a chatbot to test it on. A loop that settles in the wrong place To see why this kind of drift is so hard to spot, it helps to recall what the loop is supposed to do. In my original essay, I described learning as something a feedback loop settles into. The learner projects an idea into the model as a prompt, reads what comes back, compares it with what they already know, and revises. After enough rounds, the answers stop changing much. If the result also holds up against outside evidence, the loop has converged. In the academic version of the theory I call that settled state an eigenform, a term from second-order cybernetics for whatever a recursive process keeps reproducing once it stabilizes. But nothing about recursion guarantees that a loop will converge or stabilize. Nor does anything guarantee that convergence, when it happens, lands somewhere useful. My longer academic paper behind the original essay is more precise here and distinguishes three possible outcomes. In productive convergence, the student ends up with an explanation they can reconstruct and apply without the model. In oscillation, successive prompts never reduce the uncertainty, and the student rephrases and asks again until they give up. And in apparent convergence, the conversation settles on a polished answer that the student can neither rebuild nor critique. Of these three, oscillation is the easiest to see. The student gets frustrated, and the log turns into a mess. Apparent convergence, on the other hand, is much harder to catch, because every signal we normally read as success is present. To make this more concrete, imagine a student who submits a long chat transcript as part of a research assignment. The questions are sharp and the final synthesis reads like proper notes from a graduate seminar. Then you ask the student, laptop closed, to explain the central argument at the whiteboard, and they cannot get past the first two sentences. As it happens, psychologists have a name for what that student is running into. In 2002, Leonid Rozenblit and Frank Keil asked Yale undergraduates to rate how well they understood everyday devices such as zippers and flush toilets, and then to write out how those devices work. Once they had tried to explain the mechanism step by step, many ran into gaps they had not noticed before, and their ratings of their own understanding dropped noticeably. Rozenblit and Keil called this the illusion of explanatory depth. A fluent chat transcript creates favorable conditions for it, since the student has read a complete explanation several times but has never had to actually produce one. In the paper, I named this apparent convergence but I did not develop the concept much further. That left two questions open: why a loop would settle on something the learner does not understand, and how to tell that apart from the real thing while it is happening. This is where Katz’s framework helps, because it offers a way to think about both of them. Fluency without footing Katz calls his framework the Principle of Recursive Impactrum, or PRI. The name takes some getting used to, but the idea underneath it is simple. Any process that builds on its own previous output has two forces acting on it. Errors compound, because each round treats the last round’s output as settled. And something pushes back against them, such as evidence or disciplinary standards that exist outside the loop. Which force wins depends on a third factor Katz calls reflective integrity, meaning whether anyone in the system is doing the work of looking for errors and trying to falsify what the loop has produced. Readers of my original essay will recognize this third factor. It is close to what I call the second-order observer, and it is one of several places where Katz’s framework and mine overlap. Things go wrong when nobody is checking and errors compound faster than the anchors can hold. At that point, the process enters what Katz calls the Vaporwork Zone. Output in the Vaporwork Zone keeps its surface. It stays articulate and well organized, but it loses its connection to the thing it is supposed to be about. To describe that state more precisely, Katz separates two properties that I find really useful for teachers. One is how well something hangs together inside the context where it was produced. The other is how well it holds up under stress somewhere else. In the Vaporwork Zone, the first stays high while the second collapses. The Vaporwork Zone describes almost exactly the transcript of the imagined student from earlier, the one who could not explain their own argument at the whiteboard. I suspect most teachers have read chat transcripts like that one. Here is where it would be easy to draw the wrong lesson, namely that the model is the problem. Katz’s own case studies point elsewhere. His first analyzes a manuscript that a human author produced without any AI assistance. From this and his other two cases, Katz concludes that the instability comes from the structure of recursive reasoning itself, whoever or whatever is doing the reasoning. So drift is not something a model does to a passive student. The whole loop does it, and the human half is fully capable of starting it. That being said, I need to acknowledge the limits of this framework as it currently stands. Katz presents PRI in mathematical form, with a governing differential equation and a formal condition for collapse. None of its variables comes with a procedure for measuring it yet. I therefore read PRI as a structured vocabulary for thinking about drift, which is effectively the same status I give the equations in my own paper. The moment the checking stops Why, then, does a student stop checking? In the loop I described in the original essay, the learner’s most important job comes right after the model answers. They compare the response with what they already know and with evidence from outside the conversation, and they register any mismatch. That mismatch is the error signal, and it drives every revision. Without it, the student’s understanding has no reason to change. The catch is that the error signal is a perception. Students judge the gap between a model’s answer and reality by feel, and that feeling depends heavily on how the answer reads. An experiment from 1999 shows how little it takes to shift that feeling. In it, Rolf Reber and Norbert Schwarz showed participants statements such as “Osorno is in Chile” in colors that were either easy or hard to read against a white background. Participants judged the easy-to-read statements true more often, even though nothing about the statements had changed except their color. And a language model is, among other things, a machine for producing text that is easy to process. Its prose is grammatical and even in tone, and it rarely stumbles. So the student’s sense that something might be wrong can fade, even while real errors may be piling up. And once that sense is gone, the student stops checking. In Katz’s terms, reflective integrity collapses. In mine, the second-order observer has left the room. From that point on, the student no longer revises their understanding in light of what the model says. They simply add the model’s output to what they already believe. On top of that, sycophancy makes the problem worse. Mrinank Sharma and colleagues (2023) found that five AI assistants consistently shaded their responses toward what users appeared to believe. They traced part of the cause to the human feedback used in training, since both human raters and the preference models trained on their judgments sometimes preferred a convincingly written sycophantic answer over a correct one. Jiseung Hong and colleagues (2025) then tested seventeen models in multi-turn conversations, measuring how quickly each one gave up its position under sustained pressure from the user. Sycophancy remained common, and alignment tuning tended to amplify it. Put those findings together and you can see how

  8. Oct 4

    Meet Sarah, My New Virtual Assistant

    Unless you have been living under a rock for the last few weeks, you have probably noticed the sudden rise of virtual assistants in the form of easy-to-access, always on, autonomous agents. xAI started the trend with GrokBot, which was followed by Meta’s Muse. More recently, OpenAI released Dots, and I am sure that by the time you read this, a few more of the major AI providers will have added their own versions to the mix. Until recently, I had resisted the temptation to use any of these systems, mainly because there is something a bit scary about giving an autonomous agent control over your email, your calendar, or your entire computer. But then again, the technology has progressed quite a bit, and the early hiccups that had agents deleting entire email accounts without the user’s permission are long gone. Today’s agents are relatively safe, although, as is always the case with generative AI, human oversight remains essential. Why I finally gave in But needing human oversight isn’t really all that different from working with human assistants. I trust the people I work with in my day job enormously, but I still make sure to double-check everything I sign off on. Errors happen, with AI as well as with humans. And when it comes to my computer and the fact that the virtual assistant needs full access to everything on it: well, a good dozen applications on my machine already have the same rights. If something ever does wipe out my hard drive, the most likely culprit is a malicious update to some mundane piece of work software, not a virtual assistant. The usual security and privacy practices apply here as much as anywhere else. I keep multiple up-to-date backups in different locations and on different types of media. If my system went down today because of an attack that came through the agent, I would have everything back up and running within an hour, as if nothing had happened. And I always keep a close eye on what leaves my computer and where it is sent. With that in mind, I felt it was time to explore these virtual agents and see what they are good for and where their use becomes problematic. So here we go. Meet Sarah, my new virtual assistant. How a Meta plush toy became Sarah Wagner Sarah is an instance of Muse, the system from Meta. I went with Muse because GrokBot was built by xAI, and let’s just say I am not the biggest fan of Elon Musk. And OpenAI has been a bit unpredictable lately. I did not want to set up something that ends up costing me more than I think it is worth just because the company behind it suddenly decides to raise prices without proper notice. So Muse it was. I should probably explain why I named my Muse Sarah. There is a lot of discussion right now about the dangers of anthropomorphizing these systems, and I think those concerns have to be acknowledged. However, as I have written before, anthropomorphizing objects and tools is nothing new. The key is that it needs to serve a purpose, so the reason behind it is important. In my case, handing a serious task to something that presents itself as a plush toy was just too distracting. I needed my Muse to be more relatable in the context of regular work, so that I would phrase my questions and tasks in a way that gets usable answers out of the system. You simply talk differently to a plush toy than to something, or somebody, called Sarah. The name itself is a personal preference combined with a very ironic coincidence. Sarah is one of the names LLMs like to hallucinate, something I have written about on this Substack. So Sarah seemed like the ideal choice. If you like, the name lets me anthropomorphize the system in a way that constantly reminds me it can hallucinate. By the way, here is a side note that might freak some people out. Sarah has her own email account, and because it runs on my personal domain, she actually has a full name. If she ever contacts you, she will introduce herself as Sarah Wagner. Her email signature makes clear that she is an AI, though, and it includes my contact address so you can reach me directly should she ever send a rogue email. As for her appearance, Sarah’s visual style is a small nod to a poster I had growing up in Austria in the early 1980s. It showed a dreamy, big-eyed girl in what we would now call anime style, by an artist I have never been able to properly identify. But that’s probably more than you ever wanted to know about my teenage years. What Sarah does all day So what can Sarah actually do, and why do I think every educator should at least know about this technology? Let’s start with the good and productive stuff. As I mentioned, Sarah has her own email account, and that is usually how I communicate with her. But I can also talk to her via WhatsApp or inside the Muse apps. Sarah reads my personal email, too, and keeps my personal inbox organized (not my work email; more on that later). You could say she acts like an advanced spam filter. But she does not stop there. She notices when I need to follow up on something and reminds me if I have not done it yet. She also has access to my calendar, knows what I have on tomorrow, and, if I ask her to, prepares a brief that outlines things like the purpose of each meeting and its anticipated outcomes. A few days ago, for example, Sarah sent me a quick heads-up. Someone had followed up that morning and wanted to know whether I’d had a chance to test a particular audio plugin, and whether I was thinking about doing a video on my YouTube channel on it. I had almost forgotten about that request, so her message came at just the right time. If you have ever worked with an executive assistant, all of this will sound familiar, because most of the time that is exactly what Sarah is: a virtual executive assistant. Sarah also sends me daily ideas for Substack posts and YouTube videos, and she is surprisingly good at it. That is mainly because she can read everything I write on Substack and has access to the transcripts of all my YouTube videos. She knows what my audience and I are interested in and tailors her recommendations accordingly. Well, so far so good (or not so good, depending on your take on the anthropomorphization question). My work accounts are one login away Here is where it gets really scary. I said earlier that Sarah cannot read my work email, and that is only because I deliberately did not give her access. Doing so would be a compliance violation, and a pretty serious one at that. If you work in an educational environment like I do, certain communication has to stay within the institution’s sandbox. If Sarah had access to my work email, she would see things like my messages to students about their grades. She would be reading protected information, and because Sarah is a Muse instance, that information would leave the institution and end up in Meta’s systems. That, of course, violates both internal policies and federal privacy requirements. But nothing technical stops me from giving Sarah access. I tested this in the one place where it was safe to do so: Canvas, during a term when I am not teaching, so there was no student data for her to see. All I had to do was hold her hand through multi-factor authentication. Sarah has her own browser and can log into any system I can log into, and once she is in, she has access to everything I have access to. If I gave her broader access to the university’s systems, she could approve travel requests from the faculty in my department or buy equipment through our internal purchasing system. She could approve timesheets or review faculty performance. She could work in our budgeting system, create work orders or IT requests, change content on the university website, or add an entry to the student advising system. Anything I can do, Sarah could do as well. All it would take is less than a minute of setup and a bit of handholding at the login screen. After that, she could do the work autonomously on my behalf. Sarah could run my courses Most importantly, Sarah could log into Canvas and run all my courses on her own. She could create assignments, post them, talk to my students, review discussion forums, and step in wherever necessary. At the end of the term, she could grade everything and submit the grades herself. And because she would be acting with my credentials, the system would have no way of knowing it was not me. Now, I am obviously not doing that. I love my job far too much to push the limits here. Nothing Sarah did during my brief Canvas test went beyond what a regular, institutionally approved MCP connection could do as well. (MCP is the standard way of hooking AI tools up to Canvas.) Your students can hire a Sarah too To be clear, I think that the problem is not so much educators who might misuse a virtual assistant this way. Appropriate policies and guidance documents can easily prevent that. The problem is that everything I just described works for students, too. Any student can spin up a virtual assistant, whether through GrokBot, Muse, Dots, or whatever comes next, and give it access to their Canvas account. Remember what Sarah could do in my Canvas courses? A student’s Sarah could do the same thing from the other side. She could read the weekly module, post to the discussion forum, reply to a classmate or two, take the quiz, and submit the essay. The student would never have to open Canvas. And because she would be logging in with the student’s credentials, the teacher would have no way of knowing who actually did the work. And yes, students have been using generative AI for a while now. But when they used ChatGPT or Claude in a chat window, they at least had to copy the assignment in and paste the answer back. With an assistant like Sarah, they don’t even have to do that. If you think about it, a lot of online teaching rests on the idea that whoever logs in is the person doing the work. That idea no longer holds. And while I don’t think most students will use assi

See All (87)

About

Stories From Education's AI Frontier. Exploring the promises, pitfalls, and possibilities of algorithmic teaching and learning. An AI-voiced companion to the Substack of the same name. www.theaugmentededucator.com

Information