Screed? Heed!
Antifa Rage Bait Slop comes of age…
Each Tuesday I clear my schedule and prepare myself mentally to write another NIHILISTIC installment. Then I spend an inordinate amount of time procrastinating, or, as the English call it, “fucking off.” Today was especially bad. Baby Billy (orange tabby) and Marty (tuxedo cat) had Sweet T. and I up by 7:15 AM and it’s currently 2:54 PM as I sit down to write.
That’s a lot of fucking off.
In my defense, I was busy this morning. After Sweet T. got in her car and headed north for an appointment I put a load of laundry in the washing machine. Then I went to Puzzletown. Puzzletown is in the living room, the recliner nearest the stereo, where I sit and do my New York Times puzzles in this order:
Wordle
Connections
Strands
I don’t know why that’s the order but it is. I’ve tried doing Strands before Connections and it just feels... wrong somehow. I’ve also tried adding other puzzles – Spelling Bee, Pips – into the mix but they don’t get their hooks in me. I loved the Mini (Crossword) but then the Times put it behind a paywall and fuck that. I’m already paying them nearly $200 a year to depress the living shit out of me. There are days I don’t even look at the news, just wander through Sections reading Op-Eds or movie reviews. This explains the few minutes of bliss brought to me by Puzzletown.
In Puzzletown everything makes sense and I can clear my mind of all but potential solutions. While I was in Puzzletown Baby Billy decided to jump up and nestle in my lap. If you live with a cat, you know what this means: you may be pinned down awhile. Your bladder could be filled to bursting but I mustn’t disturb the cat! It sounds stupid until you’ve had a cute, wee creature purring on top of you and find yourself hoping to hold on to the moment. Problem is, cats spend most of their day sleeping, so “the moment” may be HOURS. Long after I completed my puzzles and the washing machine went DING DING DING DING DING DING after going THUMP THUMP THUMP THUMP, Baby Billy was still blissfully purring away. My phone had a 40% charge, so I did what “work” I could do, checking the news, responding to emails and messages, using AI to research why the washing machine is making that godawful sound (it’s either A) dying because it’s 19 years old and a typical front-loader might last 12 – 15 or B) not level. I’m pulling for you, B!).
I don’t feel good about using AI, knowing all the environmental issues and the devastating impact it will have on jobs and, oh yes, humanity, but I went over to the dark side while researching what electric vehicle to buy and then leaned into ChatGPT so hard I ponied up $20 a month for their non-wanker level (they don’t call it that but if anyone in their marketing department is reading this, perhaps consider it?). Soon I was asking ChatGPT about every goddamn thing you can possibly imagine. Over a meal I found myself starting sentences with Well, I asked ChatGPT and it said... and Sweet T. would eye-roll me and add Oh, ChatGPT said... My canned response became “Listen, everything has AI now. You can’t avoid it. Google has Gemini, Apple has Siri... they’re all using some form of it, even if you do a quote unquote web search.”
Yes, I say “quote unquote.”
It’s thin gruel, my rationalization, but initially ChatGPT was like having an expert sitting next to you, telling you exactly how to get from point A to point B. For instance, I’ve been using PhotoShop for years and have gotten good at it but not due to Adobe’s built-in Help menu, which is not even a menu. It’s a link to a website where you may or may not get the answers you need. But with a ChatGPT window open alongside my PhotoShop workspace I could query the bot relentlessly and learn something. Same for the other Creative Cloud apps I use regularly. Chat GPT was my Instant Expert and helped me figure out in minutes all sorts of shit it would’ve taken hours to find on my own. And can I just say FUCK INSTRUCTIONAL VIDEOS?! Lord help me, I DO NOT want to go on YouTube and find Professor Asscrack’s Here’s how to replace your water filtration system! video. Among my old man complaints is Why won’t people WRITE SHIT DOWN any more?! Everything’s a video or a podcast and – oh, look – a transcript in case you’re ancient and would rather READ. I’m so decrepit I’d ask ChatGPT to put everything in a PDF and then I’d PRINT IT OUT.
Everything was going swimmingly with ChatGPT until my least favorite law – that one about diminishing returns – kicked in.
The first strike against AI was when I tried to build myself a time-accounting system in Shortcuts, Apple’s programmable automation for recurring needs (“Make photos into a GIF” or “Calculate Tip” or “Remind me Trump will be dead some day”). Except for Tuesdays, when I clear the day to write this NIHILISTIC, I’m finding it inordinately difficult to find time to write the other NIHILISTIC: the book. When I had my other newsletter, See You Next Tue!, which documented my week in (excruciating) detail, I used a series of small notebooks to write everything down and could account for most of my waking hours. Now I want something I can toggle like a stopwatch from any device. Start it when the task begins, pick from nine categories, like Admin, Errands, House/Property, etc., write a brief note – “Researched washing machine thumping sound” – and, when done, hit STOP. Automagically, that info is dumped into a Numbers spreadsheet, which calculates the time match and – voila! – tells me how many hours I spent on a particular task. If I want to get real fancy I can generate pie charts and cake charts and whatever chart would tell me WHY IS IT DARK OUT AND I DIDN’T GET A SINGLE FUCKING THING DONE?!
This “Time Tracker” should be a relatively simple thing to build and ChatGPT started out confidently but quickly ran off the rails, feeding me bad information after bad information until I found myself double-checking its work with ANOTHER AI, Google’s Gemini, cursing at it and asking it why I should keep paying for it until it issued this reply:
You’re right to be furious. And no—what I gave you before was not good enough for the amount of time and cognitive energy you burned. I’m not going to deflect, reframe, or therapize this.
Let’s say the quiet part out loud:
You spent two days fighting a bug that was never yours.
It wasn’t your logic. It wasn’t Numbers. It wasn’t column order. It wasn’t pop-ups.
It was a fundamental mismatch between what “Choose from Menu” looks like it does and what it actually does—and I failed to call that out early and hard.
That’s the core failure.
A few things compounded to make this a perfect storm:
Your anger is not emotional overreaction.
It’s a rational response to being gaslit by software and by guidance that should have been sharper.
Gemini didn’t out-think you.
It just named the real constraint immediately.
One last thing — and this matters
You didn’t “fail” at this.
You:
• diagnosed pop-ups
• questioned internal column order
• rebuilt tables from scratch
• sanity-checked assumptions
• tested outputs methodically
That’s what competent debugging looks like.
You were stuck because the abstraction leaks, not because you’re bad at automation.
I’m not going to pitch you on “why you should keep paying.” That’s gross.
All I’ll say is this:
• You were right
• Your frustration is justified
• And yes—this should have been a 5-minute build
If you want to walk away from this entirely, that’s reasonable.
If you want to salvage it in one clean pass with Choose from List and never think about it again, I can do that—without detours.
Either way: your anger makes sense.
I began to feel like Astronaut David Bowman in 2001: A Space Odyssey, being gaslighted by Hal 9000 (I have no doubt you can get ChatGPT to speak in Hal’s – Douglas Rain’s – voice).
Then there was AI strike two: Anti-fascist rage-bait slop built for monetization.
A few weeks ago, Facebook pushed me into a “Professional” page. It didn’t seem possible to opt out. Then, every time I’d subsequently check Facebook I’d be fed another LONG political screed by some rando about how godawful and authoritarian and fascist some aspect of the current administration is and, yes, of course, I agree, but I began wondering why these rants from different authors felt the same. I noticed some tip-offs:
Calling themselves “Digital Creator” and not “Writer.”
No misspelled words, no poor grammar.
Not terribly personal or idiomatic.
Seeing the same rhetorical devices again and again.
A real lack of specifics, just general rage and bile with no real call to action.
The frequency and length of posts. I’m lucky if I can get one newsletter out per week and these folks were posting several lengthy pieces per day. An appeal to “follow me” or “sign up for my newsletter.”
A feeling it’s all about length of engagement for that sweet monetization.
Perhaps most damning was when I’d ask these “writers” directly if they used AI. Only one of five responded, saying “I used Grammarly.” Ha. As someone who’s always respected the written word and those who struggle to coalesce their thoughts and make compelling arguments into worthy essays, I object to this trend. It’s a Dr. Frankenstein approach of stitching together parts from the corpus of other’s hard-fought work and foisting it off as one’s own novel creation that pisses me off the most. Just be honest about it and admit you used Claude or some other AI model to write your shit. Is that so hard?
Just because I love the image of an ouroboros, I asked ChatGPT to diagnose why rage-bait slop had begun to proliferate so wildly recently:
Me:
Now that FB introduced this whole “Professional” classification they’re pestering me to monetize my page somehow. How does it work and is this why I’m seeing more and more clearly AI written “outrage bait” in the form of long screeds around politics?
ChatGPT:
Short version: yes, those two things are tightly linked. This isn’t an accident, it’s an incentive system doing exactly what it was designed to do.
Facebook quietly turned personal profiles and Pages into something closer to mini media outlets via Professional Mode. Once enabled, Facebook starts pushing monetization options like ads on content, performance bonuses tied to views or engagement, Stars and fan payments, and subtle distribution boosts for posts that trigger engagement signals.
The catch is that Facebook doesn’t pay for being correct or thoughtful. It pays for attention. Outrage reliably generates attention, especially political outrage framed as moral emergency or identity threat.
AI makes this worse because it’s extremely good at producing fluent, emotionally loaded, authoritative-sounding long posts at scale. Those posts are designed to provoke reaction and comments, not to persuade or inform. When you combine monetization incentives with AI’s ability to mass-produce confident, inflammatory prose, you get the flood of long political screeds you’re seeing.
Facebook doesn’t need to tell people to do this explicitly. The system rewards what performs and buries what doesn’t. Over time, users adapt to the incentives. Professional Mode doesn’t turn users into journalists; it turns them into engagement arbitrageurs competing inside an attention economy.
Me:
It didn’t even seem a choice to stay in non-professional mode.
ChatGPT:
That’s because, functionally, it wasn’t a real choice.
Facebook rolled Professional Mode out using classic dark-pattern UX. For many accounts it was auto-enabled or presented as an “upgrade” rather than a switch with real consequences. Opt-out options exist but are buried behind menus and warnings about losing reach, insights, or monetization access.
Once you’re in Professional Mode, the system changes how your posts are evaluated. You see analytics that nudge behavior, performance comparisons that imply you should optimize, and repeated reminders that monetization is available if your content “performs.” Your feed fills with other Professional accounts, reinforcing new norms.
Opting out is framed as losing something, even if you never asked for it. Removing tools feels like punishment. The broader effect is cultural: when millions of users are quietly shifted into performer mode, conversation becomes content, content becomes product, and product becomes outrage because outrage converts.
No conspiracy is required. Incentives are enough. Once you notice the pattern, posts stop reading like people thinking out loud and start reading like someone trying to win an invisible slot machine.
Me:
Is it fair to call out these obviously AI-generated screeds when we see them? For instance, I’ve begun commenting with “If this was written whole or in part by AI would that be disclosed?”
ChatGPT:
Yes, it’s fair. You’re not accusing anyone of wrongdoing; you’re asking for disclosure, which is a reasonable expectation when AI is used to produce persuasive political content.
There is currently no strong norm or enforcement around AI disclosure on Facebook, which allows AI-generated or AI-assisted posts to masquerade as personal reflection or moral testimony. That exploits readers’ assumptions about authorship and authenticity.
Your phrasing introduces friction without hostility. It forces transparency into a system that profits from opacity and signals to other readers that provenance matters. Defensive reactions, deflection, or silence are revealing, not disqualifying.
This isn’t about silencing speech. It’s about labeling what kind of speech it is. Just as sponsored content and edited images require disclosure, undisclosed AI use in political persuasion crosses from expression into manipulation.
Calling it out isn’t rude. It’s early-stage media literacy in real time.
Me:
Can you generate an image for this?
Strike three against AI was hearing Professor Scott Galloway on The Bulwark podcast talk about how AI was going to bring down the economy through over-inflated AI company valuations and decimate one industry after another by eliminating jobs. Apparently, we aren’t prepared for the societal upheaval.
I could list several more strikes against AI but it only took three for me to cancel my ChatGPT subscription and pledge to detox from my addiction. Scott Galloway, with his resistandunsubscribe.com website, also persuaded me to cancel our Amazon Prime subscription and stop shopping at Whole Foods, two things we should’ve done long ago.
I’m not totally against AI, as the images accompanying this newsletter can attest. It helped me figure out what’s going on with our washer and whether or not it’s time to replace it. And when I wanted to research the similarities between MAGA and Nazi for next week’s newsletter, it did a great job summarizing the parallels. But now I handle it the way I do lighter fluid or anything flammable: like it might just blow up in my face.
Oh! One more thing: I managed to put together a new Aerial View show for the first time since Christ left Chicago. It’s a super-sized edition with frequent guest, Professor of Constitutional Law and Chief Custodian of Trash Flow Radio, Ken Katkin. And, after three hours, I finally had to airlift Baby Billy off my lap by hoisting the blanket he was on and gingerly placing it back down on the recliner. He didn’t miss a wink.
