A new report from Pew on how teens view and use AI says that more than half of teens turn to chatbots for help with school, and some are turning to AI for emotional support and companionship.
“A brief history on writing, a forgotten technology: how the challenges we face in today’s digital age can be traced back to one of the most significant yet underrated innovations of all time.”
“Digital silence” is a new ethic emerging on the visual web, where travelers are no longer posting about their travels. Reading along, thinking of how I rarely post about family out of respect for their privacy, and recently took several road trips without a single post.
My life is work right now, so I’ve been training my reading and writing habits in that direction in the hope it will be additive. So when a friend who works in tech suggested I pick up some Ellen Ullman, I snapped it up. Ullman was a programmer who wrote about her experience as a woman in tech in the 1990s, a diligent personal accounting of the early days of Silicon Valley that foreshadows so much of what people are worried about today. Through her first person account of life as a programmer, she consistently reminds the reader that computers are made of boxes and wires, with choices made by mortals (often imbued with dreams of immortality) written on chips and tape, and are limited to only know what we tell them. The Y2K essay was an especially welcome reminder in the era of “singularity” — we’ve been here before.
My friend Dave Bangert has been a Midwestern reporter for thirty years, and began his own newsletter project after layoffs and mergers emptied his local newsroom. Now his 8000 subscriber project is a major influence in a Big Ten town that otherwise has a weak media market. In this podcast he talks about how it all works, what we lose as legacy media dries up, and what it means when your readers know exactly who you are, and you run into them in the produce aisle.
As a true blue Hoosier, here’s my hot take on the Chicago Bears moving to Hammond: no dunking on NWI.
I’m seeing a lot of disgust for Indiana in related Bears commentary that troubles me, especially the comments about NWI as a place of blight, beyond consideration or repair, obscuring the stakes behind a regional geography of race and class. Meanwhile, I am glad that people from Chicago suburbs have found occasion to think about their neighbors in Hammond and Gary. There is life outside of Schaumburg.
Indiana’s history as a state is thick with race and class problems, perhaps best captured in the old phrase calling it the “middle finger of the South thrust into the North.” That middle finger reaches right up and touches Chicagoland. Indiana’s policy approach to this dynamic is predictably toxic, having cleaved communities like Hammond and Gary from state investments and statehouse power. Indy loves business and it loves a chance to razz Chicago. Guess why.
“The Region” in Northwest Indiana is practically and functionally the south side of Chicago. You can drive from Gary through East Chicago and Whiting straight into Pullman and Bronzeville, without much marking a state line other than some signage and an iconic White Castle. But everything on the Indiana side of the line is governed from a statehouse that rarely considers them at all until there’s a chance to further demean and isolate them. Like so, with Indiana’s current plans for redistricting that further disenfranchise Gary residents.
Does it serve to remind us that Gary was founded as a company town by the world’s first billion dollar corporation? I think so (and great article btw). The town was the subject of serious curiosity and consternation during the early days of industrialization, especially when comparing the claims of the owners and capitalists to the realities of the workers. Poet Carl Sandburg famously wrote the poem “The Mayor of Gary” to illustrate the casual venality that the ownership class felt toward the steelworkers. This dynamic is still very much alive in the minds of the people in surrounding communities, who are perpetually worried about a “Chicago influence.”
This is why we train a jaundiced eye on characterizing working class communities like they’re beyond consideration or redemption. The world isn’t a Costco. People live among the brownfields. I spent a lot of time in this area in my twenties – all my friends in The Region were somehow connected to US Steel, even still. Feel however you feel about the Bears deal, but we don’t do Gary slander in this house.
Yanking this back to the question of writing: if you want to catch an interesting news source from the area, Capital B’s Gary outlet does some amazing reporting and opinion.
The last time I went truly viral was in April of 2020 in the height of the COVID shutdown. I posted a tweet and walked away from my phone. By the time I checked back, my notifications were out of control.
The next morning, I got a message from an old friend familiar with my handle. “I think I saw your account on Good Morning America?” Hilariously, there I was, and I couldn’t even claim responsibility for the meme.
I’ve spent most of my time on the internet as a pseudonymous account, using my first name only, or using a handful of handles (incl feministe, fauxrealtho, flotisserie, paired with punny names like “Petty White” and “Frieda People,” a convention from the Tumblr years). Even as a visible personality online, I was only known by my first name and URL and/or handle. In recent years I de- and reactivated some of my socials, so some of these breadcrumbs no longer exist, but I’ll do my best.
Pseudonymity allows writers to explore complex ideas in digital spaces while protecting their identity, location, and other identifying factors, while also maintaining a throughline of identity and storytelling.
There are a lot of trade offs in using a pseudonym, especially around how to claim credit for your work. But people use them because they give you the privacy to be honest, real, weird, authentic, and to escape the creep of modern social media presence into high stakes spaces like the workplace. This dynamic was kind of the impetus behind the era of “weird Twitter,” where people using pseudonymous social handles routinely threw out funny, absurdist one-liners to impress their friends and followers, while taking a turn at the social media slot machine. Not every post or joke lands in a way that converts to numbers, but some do, and it’s fun to try.
I’ve written a little previously about how the English program I was a part of used fiber arts to illustrate the fundamentals of technical writing, but one of their other methods of teaching the internet was through board games. Games, like the internet, and much like writing, provide rules and structure for communication and engagement, but everything that happens inside the container of game board and game play is a mystery until it emerges through human interaction. So goes the internet (and to some extent, so goes AI). Games and gaming were used as a method to think through what it means to create rules of play, then let a community rip through the model – and many of the thinking and skills involved around game design apply in social digital spaces, from chat rooms and Teams channels to the open seas of the WWW. The longer you’ve been playing the slot machine, the more you get a feel for the kind of thing that will get seen and read, and who among your readership will take your content to the next level.
Why does it matter? Because understanding what “works” to make ideas travel further online, the more you can tap into it. Big Tech is under fire right now for amplifying some of the worst impulses of the internet, by cranking engagement algorithms to exploit messages that produce outrage. Yes, big emotions create virality, but so do relatability and sharability.
So back to the meme: here’s how it went down.
The meme was a list of six hypothetical celebrity households: pick one to quarantine in.
It had been circulating on Facebook, started by a Christian influencer named Savannah Locke. I encountered it deep in a Real Housewives fan group, and felt the pull of a good parlor game. In April 2020, everyone I knew was sitting at home, fretting about the COVID-19 pandemic looming over all of us, so I shared it with minimal ceremony, a couple of buddies with a slightly larger following hit retweet, and within two days the tweet was being cited by The Cut, Time, the Washington Post, CBS, and others. Know Your Meme documented it for posterity. Several outlets named me as the originator, but trust, I was only trying to delight my friends with low grade Facebook content. I couldn’t find the original meme at the time – Locke appears to have had a name change that scrambled my search. But hey, as these things go, nobody earned a dime or promoted anything weird, so no harm, no foul. Business Insider managed to credit it correctly, so a special kudos to their editor.
The core game mechanism behind the meme is forced choice: constraints generate opinions and opinions generate activity. Each “house” also represents a personality type. House 3 is chaos, House 6 is aspirational, House 5 is the one where someone is definitely cooking and someone is definitely yelling about it. The choices are arbitrary, which invites curiosity about who grouped what and why. In short, this meme offered a light conversation starter at the right time, with low stakes, high personal reveal, and endlessly discussable combinations. It also drew from existing memes and games that are popular online, like “where you sitting” in the proverbial lunch room. This one became news because of the timing and gamability, not because I was particularly clever, but hey.
What does it feel like to go so viral? It’s hilarious, strangely affirming, and also a little crazy-making. It opens the door to a whole lot of wild people and ideas on the internet, not all of them flattering or welcome. Virality is sometimes paired with incredible harassment and requires “more condolences than congratulations.” But as far as this particular experience went – hilarious, whoops, and wow. It just felt like a Lebron James, Post Malone and Jennifer Aniston hang would be a good time.
In an Internet of things (IOT), what happens when the companies behind the service model cease to be?
What to do when the biggest platforms for readers are kind of evil?
Making the rounds on social today: New study suggests that while 70% of financial firms use AI today, 80% report no impact on employment or productivity.
“Because RSS is an open protocol, and because there are so many different possible ways to follow and read the news, RSS readers ought to be a UI playground in the way Twitter clients once were, where innovation and experimentation are normal and celebrated.”
There are many hyperbolic essays on AI going viral this week. This essay gets into some of the emerging politics around AI in the United States and how they map onto electoral politics, and is irritated with the American left’s approach to AI.
I agree on one angle, that as a cohort, abstaining entirely from new tech is a bad approach.
Haraway (obvs) suggests we stick with the trouble, that sometimes our very presence in the room is what’s required to trouble existing narratives, and that it’s important that share what we learn back to our people at home, whatever that means in our context.
For me: I spend a lot of time at work translating technical ideas and projects in an institutional voice, but on nights and weekends, I’m explaining the ins and outs of the internet to my working class friends and family. Increasingly, I’m asked to explain LLMs and AI and how it relates to their needs around business, the news, entertainment, and as a legal aid.
The “cyborg’s mark” is a metaphor for writing, our duty to translate, and the inevitability of translation through experience. Perhaps staying with the trouble allows us to articulate our experiences in and around the science in ways that both advance our interests and that keep our communities safer. For those of us who straddle multiple worlds, translation is a responsibility.
Widespread AI adoption reopens some basic questions in business: who your audience is, what work can reasonably be automated, what absolutely requires human oversight, and how the service and information environment actually holds together when we shuffle these circumstances around. All this, with many social and environmental risks and a side of existential doom.
The hype makes it harder to see a change that’s already here. To me, the shift between SEO to GEO is one of the clearest places where a change in technology produces a significant process change in the relationship between a writer and an audience, with all the tradeoffs around authority, context and information architecture that implies. But web search is just one place this shows up.
I suspect the predictions of economic and social doom are overblown by a lot – in life as in business, generalisms are easy and specifics are hard. Last mile issues remain a reality of nearly any technical or software implementation, and differentiation is in many ways the art of business. There are serious incentives to promising a smooth, profitable near-future, especially around emerging markets, so I feel pretty comfortable forecasting that the era we currently live in is valuation masquerading as value. Instead, I think we should worry more about how the tech changes our relationship to information and to each other.
I’ve been writing about how writing and code are the same thing in digital environments, and about how that equivalence shaped the early web. AI changes that relationship.
The dominant conversation is about whether LLMs can write well, but I suspect that’s the wrong frame. Human storytelling will probably always be more interesting than generated storytelling, because humans love quirks and novelty that can’t be produced artificially.
The more consequential change is that AI-generated text doesn’t just sit on the web waiting to be read, and instead feeds back into the system that produced it. It becomes training data, source material, and eventually, architecture. Remember: When an LLM generates text, it’s producing word sequences based on statistical patterns. The output is one plausible version to your prompt, not a definitive one — but it gets indexed, linked, and cited like any other writing. Nothing about its surface tells you it doesn’t carry the same authority.
Researchers call what follows “model collapse,” a feedback loop where models trained on AI-generated content lose touch with the range of human-produced data. The rare and specific details disappear first, then the middle narrows. Eventually what’s left is smooth, confident, increasingly generic text that sounds authoritative whether it’s accurate or not, which becomes the training data for the next round.
I’m thinking about it in terms of the shift from SEO to GEO. SEO preserved a connection between writing and human judgment. Someone wrote content, search engines indexed it, readers got a list of links and decided which to trust by comparing to their own experience and knowledge. This system was gameable through various sleights of hand, but it assumed a reader with agency. The creator’s job was to be easy to find and worth finding. Streaming video complicated this process but still worked with the same basic ideas. Meanwhile, GEO operates on a different premise. The goal isn’t to get found by a person, but to be found by an algorithm assembling a response the user may or may not independently verify.
(Sad news: today, only about 8% of LLM users verify their output against source material.)
(This does not bode well.)
Consider what happens to the same piece of writing in each system. In the SEO world, your article gets indexed, shows up in search results, someone clicks through, reads it, evaluates whether you or your institution is credible on the topic, maybe skeets it or sends it to a colleague. A human encountered your work, weighed it and decided it was useful, the algo responds and indexes accordingly.
In GEO, an AI system parses that same article, extracts the most clearly structured claims, and drops them into a synthesized answer alongside fragments from other sources the user never sees individually. The reader gets a confident, blended paragraph.
In the old way, the reader moved through the web. AI yanks that experience into a single response from a single interface. We don’t fully understand how AI systems decide what to cite, which makes this power shift feel especially risky. Worse, different people will get different responses from LLMs, even using the same prompts and source materials. We don’t know why.
Fewer entry points to the web means fewer opportunities for diverse or unexpected sources to gain traction, which means the training data gets narrower, which means the outputs get more generic, which means the architecture narrows further, which means fewer perspectives represented in the output. For the reader, it accelerates context collapse in much the same way. Fewer inputs means fewer opportunities to stress test your ideas against new information.
So, what to do?
If generative AI grows as predicted, SEO and GEO will coexist for awhile, and working developers and communicators will need to understand both and how they layer. Strong SEO foundations give you a great head start in AI visibility too, so the fundamentals of good writing and web taxonomy still matter a lot.
But the production of knowledge, the keeping of data, and how it’s all indexed are subjects that are about to become very important, and very political. So I suspect that any fields that touch those topics will also become very important, and very political, very soon.
When I started building websites in the late ’90s, the line between writing and coding didn’t really exist. A person probably learned HTML because she had something to say and needed a place to put it. The internet was free and anonymous and it felt audacious to put your stuff online, like flinging a message in a bottle out to sea. The code was a container for the ideas that rendered them onscreen, and every post and page you published was both a piece of your thinking and a brick in something larger.
People forget that the early web was a writing community. Writers, or bloggers, built their own sites, maintained their own archives, linked to each other deliberately. A blogroll was both a reading list and a show of solidarity, a trackback was a way of saying, “I see you, I’m thinking with you.” The technical architecture - RSS feeds, permalinks, comment threads - existed to organize the writing and the writers’ thoughts, and to push their ideas forward on the open web.
This worked for a time. Communities of writers, most of them without institutional backing or media credentials, built new bodies of knowledge together through interacting as readers and writers, communicating across a foundation of code. The work didn’t stay online. It spilled into conference halls and state houses and newsrooms and policy discussions. As the body of communication built, it created something that accumulated over time. These people influenced mainstream journalism, shaped public conversations, launched careers and movements. In many ways, the national political conditions we face today are a reaction to that movement, and how it allowed regular people to influence the world through the democratization of mass communication.
Midway through the aughts, the brick and mortar publishers and venture capitalists started looking across the landscape, at all the writers creating fantastic content, largely for free, and sucked them into their content and editorial teams. Google Reader lost institutional and financial support as writers moved off the open web and onto publishing platforms, often backed by VC money, that measured the quality of your work by engagement. The addition of algorithmic feeds further broke down this structure – the algorithm doesn’t measure whether your work contributed to shared understanding, but whether it generated a click, a share. The code changed, and the writing changed with it.
The writers changed with it, too. The blogger became the influencer. Bloggers operated in a gift economy of ideas: you wrote to think, to argue, to contribute, and your standing in the wider community came from the quality of your work and contributions over time. Was it a meritocracy? No, but the conditions made it possible for a regular person to talk with experts as peers, which upended traditional power structures around authority and expertise (in both directions, good and bad). Meanwhile, influencers operate in a heavily capitalized attention economy where engagement converts to dollars. The audience is a market to press for money.
The gendered dimension of this shift matters as well. The early blogosphere was full of women writing sharp, rigorous work about politics, culture, parenthood, identity, and technology — work that was explicitly feminist and anti-racist and genuinely moved public conversations. This was the community I helped build (Feministe.us was my project, a community platform of writers and commenters whose coverage and discussion broadly fell under, but was certainly not limited to, the topic of feminism). When the monetized platforms absorbed that energy, the commercial model recast women’s online authority almost entirely in terms of consumer influence. What could we sell? And to whom? The framing around our work went from “this person has important ideas” to “this person can sell things to a niche market.” Meanwhile, men who’d built audiences through tech or political blogging were more likely to be absorbed into mainstream media as columnists and analysts, roles that kept their intellectual authority intact. The influencer label, with all its connotations of superficiality, landed disproportionately on women, and it stuck.
There’s a class piece here, too. The platform model offered something the early blogosphere mostly didn’t — a way to get paid. For women who’d been doing enormous amounts of unpaid intellectual labor building online communities, the question of monetization wasn’t shallow. The implications of information centralization and monetization were as present then as they are now with LLMs and AI. Some people figured out the social platforms and worked their way into viable digital careers. Platforms offered a lot of perks, but all of the perks had a backstop. Corporate interests introduced the problems of advertising, audience and sponsorship, which meant reorienting your individual practice around maximizing your commercial value over and above your intellectual contribution and community management skills. It often meant giving away some or all of your IP rights.
For most people, new system didn’t offer a viable way to get from “respected independent writer” to “respected, protected and compensated writer.” Many of us found ourselves in positions too precarious to take the leap into freelancing and social media, and some, like me, got regular jobs doing regular stuff. Some married money. And in the meantime, some folks figured out how to get into real journalism, which looks much different in 2026.
Great storytelling helps people understand themselves and their world. We let some of that depth go on the Internet with the onslaught of digital marketing and all of its implications, and today the internet feels less useful and less trustworthy than it once did.
It feels like there’s something to take forward from the experience.
I’ve been running a series of informal experiments with my team to understand how we should advise people on using AI tools like ChatGPT, Claude and Gemini around our work.
In our last session, I gave everyone the same task: ask your AI tool to summarize a single article in plain language. Every person got a factually accurate summary, but the style and depth of analysis varied widely from person to person. We tested this further by entering the same prompt multiple times. Each time, the output was slightly different. We copied and pasted everything into a shared document so we could compare side by side.
Later that day, I asked everyone to reopen the tool they’d used that morning and repeat the prompt that had worked best: summarize this article in plain language. The outputs shifted again. Some changed dramatically. One team member whose morning summary had been light and conversational received a much more formal, serious version in the afternoon.
I then asked Claude to compare the morning and afternoon versions of one summary. “In short, the morning version emphasizes practical implications and accessibility; the afternoon version prioritizes completeness and specificity. They complement each other well — someone reading both would get a fuller picture than either one alone provides.”
Why would Claude produce these differences despite the same source material and prompt?
I asked the source.
Even with the same prompt and source material, LLMs don't produce identical outputs each time. This is by design — there's a degree of randomness (called "temperature") in how the model selects words, which means each run produces a slightly different path through the text.But the differences here go beyond random word choices. The morning version landed on a journalistic structure while the afternoon version landed on an academic one. That likely reflects what the paper calls sensitivity to framing and order effects — one of the cognitive biases it catalogs. Small, invisible differences in how the model’s internal processing unfolds at the start of generation can cascade into meaningfully different organizational choices. Once the morning version opened with a news-style framing, the rest followed that pattern. Once the afternoon version opened with a descriptive academic tone, it followed that thread instead.
This is actually a concrete example of the robustness issues the paper identifies. The underlying reasoning about the paper’s content is consistent across both summaries — neither gets the facts wrong. But the presentation varies in ways that weren’t requested, which means someone relying on just one version would get a subtly incomplete picture depending on which run they happened to see.
It’s a useful reminder that LLM outputs are one plausible version of a response, not the definitive one.
It’s important to understand how these tools work. What looks like an authoritative answer is really just one of many possible outputs the system could have generated. The text will sound plausible whether it’s accurate or not, and you may get different results each time you ask.
I asked Claude.ai for a plain language summary of this buzzy new research paper. Claude offers the following:
Researchers from Caltech and Stanford conducted the first comprehensive survey of how and why large language models (LLMs) — AI systems like ChatGPT or Claude — fail at reasoning tasks. Rather than focusing on what AI can do well, they cataloged and organized the ways it goes wrong.Why it matters
LLMs are increasingly used in high-stakes settings, but their reasoning failures are often unpredictable and poorly understood. The researchers argue that systematically studying failures — like accident investigations in aviation — is essential for building more reliable AI.The researchers created a framework with two dimensions:
• Type of reasoning: informal (intuitive, social), formal (logic, math), and embodied (physical, spatial)
• Type of failure: fundamental flaws built into the architecture; domain-specific weaknesses; and robustness problems (inconsistent performance when small details change)Informal/intuitive reasoning
LLMs exhibit human-like cognitive biases — confirmation bias, anchoring, framing effects — but without human ability to recognize and correct for them. They also struggle with “theory of mind” (understanding what others believe or intend), and with applying consistent moral or ethical reasoning.Formal/logical reasoning
LLMs often can’t reverse simple logical relationships (if they know “A is B,” they may not infer “B is A”). They struggle to chain multiple reasoning steps together. Basic counting and arithmetic fail in ways that seem surprising given their other capabilities.Embodied/physical reasoning
LLMs have poor intuitions about the physical world — gravity, spatial relationships, object properties — because they’ve learned only from text, not from physical experience. This extends to visual AI systems as well.Many failures trace back to how LLMs are trained: they predict the next word in a sequence rather than reasoning deliberately. This makes them good at pattern-matching but unreliable when tasks require genuine logical inference, especially under slight variations in how a question is phrased.
Researchers have proposed fixes including better training data, techniques that force step-by-step reasoning (like “chain-of-thought” prompting), connecting LLMs to external tools like calculators or physics simulators, and architectural changes. However, no single fix is comprehensive — many improvements in one area don’t transfer to others.