I used to run into this wall all the time. I call it the Blank Page of Doom. You know the feeling: you’ve got ideas swirling in your head, but the moment you sit down to write, they scatter like cockroaches when you flip on the light. You stare at that blinking cursor, and the dread just builds. It feels impossible before you even start.
Then I read this awesome story from a guy named Gary Pratt. He’s 70, and he found the most unexpected partner to cure his own Blank Page Syndrome: an AI chatbot. He didn’t ask it to write for him. Instead, he used it as a sparring partner. He’d throw his messy ideas at it, and the AI would throw back questions, phrases, and suggestions. He said the magic happened when he disagreed with the AI, because that tension forced him to figure out what he really wanted to say. The AI became a reflective surface that helped him hear his own thoughts.
How cool is that? It wasn’t about replacing his creativity; it was about unlocking it. For a minute, I thought, “Okay, this is it. This is the perfect use case for AI.”
But then I kept reading, and a much more complicated, and honestly, scarier picture started to emerge. It turns out, AI isn’t just a friendly writing buddy. It’s creating some massive new problems we need to talk about, like, yesterday.
🧠 The Coming “Cognitive Divide”
This is the one that really got me. A writer named Nitesh Kumar pointed out something chilling:
AI isn’t just automating labor, it’s automating thinking. And it’s creating a new social class system based on who gets to do it.
Think about it. Entry-level jobs used to be our “cognitive apprenticeships.” When you started as a junior paralegal, a new consultant, or a rookie analyst, you didn’t just fetch coffee. You learned how to think. You were paid to learn how to:
- Synthesize mountains of information.
- Build a logical argument.
- Analyze data and spot patterns.
- Make judgment calls.
- Learn from failure and revise your work.
These roles were the training ground for developing a critical mind. Now, AI is coming for them. As one editorial put it, those jobs are the easiest to replicate. Paralegals are being replaced by search bots. Memos are drafted by chatbots. A writer named Will Vaughan made the brutal point that this is a zero-sum game. If AI takes a job to save a company money, that job is just… gone. There’s no replacement. The company wins, and the entry-level human loses their shot.
So what happens next? Kumar argues we get a “cognitive divide.”
On one side, you have the people who design, direct, and question the AI systems. They get to live in a world of ambiguity, judgment, and deep thought.
On the other side, you have everyone else, whose intellectual work is flattened, summarized, or just bypassed entirely. It’s a world where thinking deeply isn’t just unnecessary, it’s a skill you might never get the chance to practice. Critical reasoning and original thought could become luxury skills, hoarded by a few while the rest of us just… prompt.
⚙️ The Garbage In, Garbage Out Apocalypse
Okay, so where are these AI models getting their smarts? From us. Or more accurately, from everything we’ve ever put on the internet.
One writer, Glenn Conway, put it perfectly:
AI companies are “robbing blind” places like Wikipedia and news sites. They’re scraping terabytes of data that people spent years creating, curating, and funding, and they’re getting it for free. They vacuum it all up to train their models, then turn around and sell access to it. It’s a business model built on a free lunch that someone else paid for.
But here’s where it gets even worse. What if the stuff they’re scraping is… slop? Douglas Coe highlighted a report showing how Russian disinformation networks are flooding the internet with AI-generated garbage. This junk content is then scraped up and fed right back into the training data for the next generation of chatbots.
It’s a terrifying feedback loop. Fake content is used to train AI, which then presents that fake content as fact, making it impossible to tell what’s real. We’re dangerously close to a world where even fact-checking is impossible because the tools we’d use to check are polluted by the same lies. The only defense? As Coe says, we have to become incredibly selective and rely only on reputable news sources with a long history of actual journalism.
✨ The Productivity Myth and the Power Problem
We’ve heard this promise before, haven’t we? “This new technology will save you so much time!”
A writer named Larrie Ferreiro had a fantastic historical take on this. He pointed to vacuum cleaners from the early 1900s. They were supposed to be laborsaving, but they just created higher expectations for cleanliness. People ended up spending the same amount of time on housework. The office computer was supposed to make us all hyper-efficient, but it brought its own endless distractions and complexities.
He suggests a new rule, a spin on Parkinson’s Law:
“Work expands to fill the intelligence available, human or artificial.”
Instead of replacing us, AI will probably just lead to bosses asking us to do more. The work won’t go away; it will just get bigger and more demanding to match the new tool’s capacity. It’s not a solution, just a new kind of problem.
And speaking of problems, who gets to decide what the AI is used for? Vincent Canzoneri laid out a scary hypothetical. Imagine a government official using an AI to scan all internal communications for “disloyal” activity. An innocent email between FBI agents trying to make sure the president’s promises are being kept could be flagged by the AI as an attempt to “undermine public confidence.” The AI’s conclusion would depend entirely on the biases of the person who deployed it. It’s a recipe for automated tyranny.
🚀 So, What Do We Do? How to Use AI Without Losing Your Mind
Look, AI isn’t going anywhere. It’s a genuinely powerful tool. The question isn’t whether we should use it, but how. We have to be intentional.
We need to be more like Gary Pratt, the 70-year-old writer who used AI as a creative partner, not a ghostwriter. The goal is to stay in the driver’s seat: to be the thinker, the strategist, and the final judge. The AI is your intern, not your boss.
Here’s my game plan for using AI smartly:
- 📌 Use it as a Brainstorming Buddy: Stuck for ideas? Ask it for ten crazy angles on your topic. Use it to get the ball rolling, not to write the final shot.
- 📌 Argue With It: This is Gary’s genius move. If you’re unsure about your position, tell the AI to take the opposite stance. Defending your view against a bot is an amazing way to sharpen your own thinking and find your true voice.
- 📌 Never Trust, Always Verify: Treat everything an AI tells you as a rumor. It can and will make stuff up (they call it “hallucinating”). If it gives you a fact, a statistic, or a quote, your job is to go find the original, reputable source.
- 📌 Outsource Tasks, Not Thinking: It’s great for tedious tasks like reformatting text, summarizing a document you’ve already read, or catching typos. It’s terrible for tasks that require judgment, context, or critical thought. Don’t let it do the thinking for you.
- 📌 Be the Curator: In a world flooded with AI slop, your ability to find, vet, and rely on trustworthy, human-made sources is your new superpower. Follow the real journalists and experts.
AI is a game-changer, for better and for worse. It can be a tool that sparks our best ideas or one that makes us forget how to think at all. The choice, for now, is still ours.
- A Creative Equalizer: Research into AI’s role in creative writing suggests it can be a significant imagination booster, especially for less experienced authors, helping to narrow the skill gap between novice and seasoned writers.
- An Educational Shift: In journalism education, figures like Professor Cindy Royal of Texas State University advocate for teaching students to use AI as a tool to extend their capabilities. The focus is on augmenting human intellect, not replacing it.
- Industry-Wide Caution: Major media outlets are treading carefully. The New York Times has established strict guidelines for its staff regarding the use of AI in reporting, prioritizing the preservation of journalistic integrity and reader trust over rapid adoption.
- The Detection Dilemma: A major hurdle in managing AI’s impact is the dual threat of AI-generated misinformation and the unreliability of tools designed to detect it, placing a greater burden on human creators and consumers to be vigilant.