AI hasn’t taken my writing job — but it is living rent-free in my head

When a growing chorus of readers and critics accused Jamir Nazir, winner of the prestigious Commonwealth Short Story Prize, of using AI to write his winning entry, I did what any writer with a healthy amount of professional dread would do — I read it.

Nazir’s work, “The Serpent in the Grove,” certainly bears many of the characteristics we’ve come to associate with AI writing. Bizarre metaphors and overly compressed language initially come across as bold, but a second read — and the strange language demands one — reveals the words for what they are: eye-catching tissue paper wrapped around an empty box.

The question of who wrote “The Serpent in the Grove” was one of several AI-generated writing controversies to emerge in May — a book reviewer for The New York Times and a Nobel Prize-winning novelist were at the centers of two others. Soon after the Commonwealth Foundation announced that Nazir’s story was a regional winner, a wave of response pieces followed, most arguing that AI writing was easy to spot and that anyone who relied on it would eventually face Nazir’s fate — even though the Commonwealth Foundation’s investigation concluded that he didn’t use AI, his name and any future work will always be attached to the accusations. 

I mostly agreed that the story read like something AI would write. But I also noticed that the voices accusing Nazir tended to belong to writers, a population with an obvious stake in believing, and insisting, that an AI could never do what they did.

Those who want a more objective opinion point to AI detectors, tools that analyze text for patterns, structures, and other elements common in AI-generated writing. Pangram, the industry-leading detector, flagged Nazir’s text as 100% AI-generated. Curious to see how these tools worked — and how accurate they are — I entered an article I’d written months earlier into Pangram’s prompt box, pushing aside the irony of using AI to detect AI. Sure of my innocence, I clicked “check.” But as the tool did its thing, a thready nervousness crept into my chest. I knew I hadn’t used AI. What did I have to worry about?

A few milliseconds of unpleasant anticipation later, the screen loaded with a wash of green highlighter over my text. Phew. Then I read the score: 97% human-generated.

My anxiety curdled into indignation. I flung my cursor down the screen, furiously scrolling through the piece in search of that 3%, as if the software could sense my outrage. And there it was. My closing paragraph, highlighted in pale yellow, with a note attached: “Light AI Assistance.”

Rereading the paragraph in question, my steam began to cool off and settle into a pool of defeat and self-pity. I decidedly had not relied on AI assistance, but I could see why Pangram flagged the text: I’d written a list of three and couched it in em dashes — two common AI “tells.”

The thing is, I really liked that article, and I particularly liked how that ending came to be — born from the kind of human editorial process at the heart of most good writing. Originally, I’d submitted the piece with a different ending. My editor suggested I add a few sentences to tie back to open questions I posed in the introduction. It was a good idea, and a lightbulb flickered on as I began drafting options. She liked what I came up with, I liked what I came up with, and the main subject of the article, a scientist leading a new body of research, used it as a pull quote when he shared the piece. Readers commented and emailed to say they liked it, too. According to every standard I understood, I’d done good work.

Except now it read like something AI might write. Doubt crept into my mind. Was it really a good ending? Or was the point too tidy, too cliché, too, well, robotic?

A few months ago, I would have brushed aside the entire premise of using an AI detector as absurd, even comical — something I might try out of curiosity, but not read too much into. But by the time I was facing that yellow paragraph, I was mentally exhausted, stumbling off an edge I’d been riding all year, trying to reconcile AI’s existence with both my professional life and my identity as someone who writes. The rise of AI detectors was just another twist I couldn’t find footing on, another factor clouding my ability to read my situation, let alone anyone else’s.

My initial worry about AI centered on practical obsolescence: whether a client would literally replace me — Jasna Hodžić, a flesh-and-blood human being, one who grows tired and distracted — with a chatbot. Now that fear seemed almost quaint, a sign of a simpler time when AI only threatened my external world.

At some point over the past year, sitting down alone to work no longer meant sitting down alone to work. Even with the AI tools off, the tabs closed, I was, in some ways, considering the machine. That’s when I began to reach for the white flag. AI, I surrender: Just get me out of here, out of this place where you’re all over my process and my voice — even when you’re not.

Will write (or chat with AI) for food

I didn’t always know I wanted to be a writer. My writing habit developed the way someone with a nutritional deficiency might start craving the foods that would correct it. It was a form of self-medication, a way to calm my restless mind, which constantly swerves between several trains of thought. Putting what I was thinking into words, usually bound in a journal, forced me to slow down and focus on one track at a time, only to discover they often intersected in surprising ways.

Professionally, I pursued ecology, drawn to the field because I was spending more and more time outside and wanted to understand the world around me. I followed that field to a Ph.D. because I found the subject to be intellectually interesting. Ecology is a messy science; we tease signals from noise, but we also accept a level of unexplained variance that experts in many other sciences would find anathema. Part of being an ecologist is testing for patterns while accepting that the irreducible complexity of the natural world may always leave unsolved questions. I turned out to be good at that kind of thinking.

My favorite parts of graduate school were the opportunities to reflect on my research — my process, my discoveries — and consolidate it into something I could cleanly communicate to others, usually through writing. When several postdocs didn’t pan out, I began editing for an academic journal and then writing science stories. Eventually, I stopped applying for academic positions altogether. 

I’ve never had a byline in The New York Times, and every short story I’ve ever sent to a literary magazine has been rejected. Still, by the time ChatGPT, Claude, and the rest of the LLM chatbots arrived, I was making a decent living as a working writer. It’s true that for every piece that was meaningful — every sentence I was proud of — I had to accept several less glamorous gigs to keep the lights on. When one of those gigs ended last fall, I took it in stride. My portfolio was solid. Surely I’d find more work, maybe even a steady job.

I was familiar with AI when I started looking for work in late 2025. I’d already used ChatGPT for SEO research and had even edited AI-generated copy at a social media agency. I tolerated AI in these contexts, but I wasn’t exactly inspired to unleash the tool on my writing. At the same time, the handful of steady freelance gigs I had landed before LLMs went mainstream insulated me from the extent to which the industry had begun to revolve around AI — not just as a tool more people were using, but as a force that would commodify writing, bringing opportunity for some, obsolescence for others. 

Looking for a writing job today, in the summer of 2026, is, in a single word, demoralizing. After applying for nearly a hundred positions — dozens of which I was a good match for — I only landed a few interviews that went nowhere. I didn’t think AI was the most powerful force shaping the job market, but I also couldn’t prove it wasn’t playing a role. The possibility that someone could automate my work hovered over every rejection like a professional dementor.

Trying to combat this uncertainty with facts didn’t help. Every argument that AI posed a significant threat to working writers had an equally plausible rebuttal. For every writer I found swearing off AI, there was another who proudly admitted that they used it. Studies showing that readers disliked AI writing built up my hope that I was still relevant. That hope collapsed when blind evaluations found that readers often preferred AI writing when they didn’t know it was AI. The prevailing wisdom was that AI was useful for nonfiction — not reassuring since that kind of writing pays my bills — but hopeless at any writing rooted in context, consciousness, lived experience, or physical existence. Nazir’s story winning the Commonwealth Prize poked holes in that argument. 

Underneath every claim sat the same difficult-to-disprove assumption: AI will only get better.

Eventually, I grew cynical, and anything that argued in favor of working writers started to sound like cope — if you say something loud enough, for long enough, surely it will come true. I caught myself fantasizing about a completely different life: waiting tables, serving coffee, interacting with real people, with no ambiguity about who, or what, was on the other side of the exchange. 

Then I reached a different conclusion, one that had helped me through hard times before: I was being a drama queen. There was still a distinct possibility that I was inflating a complicated industry shift into an existential identity crisis. Is this just what it means to be a mid-career freelance writer living through yet another technological shift, only this one felt more consequential simply because it was potentially affecting my ability to find work? 

In other words, was this a case of “It’s not you, AI; it’s me”?

Just kidding. It is AI.

While AI merely hovered over me like a dementor — an ominous potential threat I couldn’t quite pin down — it had already begun baring its teeth in the lives of other writers. 

“I lost my original gig to AI, but now, in my new gig, I have to prove to AI that I’m not AI,” Rebecca Duras told me. She was funding her creative writing pursuits with content-agency work, the kind she describes as “very vulnerable” to automation. When she found a new agency explicitly seeking human writers, she thought she was back in the game. 

Her experience, alongside Nazir’s, exemplifies the difference between AI and previous technologies that disrupted the writing industry. The rise of digital media, for example, simultaneously narrowed traditional career paths and expanded the job description: build a social media presence, cultivate a Substack following, become part marketer, part influencer. That shift primarily affected the economics of writing: how writers found an audience and how they distributed, published, or monetized their writing. AI is unique. It can perform the core task of writing itself, giving employers a choice: AI, human, or a mix? Many will choose AI because it’s cheaper, but for others, like Duras’ employer, human writing carries a premium — if not for the craft and skill itself, then because human writing buffers them from potential backlash. 

But that raises the question: If human-written content now carries a premium in some circles the way a hand-built cabinet might over something from IKEA in others, how does an employer actually verify that they’re getting what they pay for?

The answer: by asking an AI to evaluate it. 

“My agency had to put everything through an ‘AI report’ for their clients,” Duras said. “And it started throwing up false positives. Every single word gets highlighted as AI, and then I had to rewrite the entire thing so it gets to an acceptable AI level.”

Duras described this job, which is far from her passion, as “annoying,” but nothing that would provoke the mini-crisis I had when Pangram flagged my writing. It also cuts into her bottom line: “It’s double the work for me because I’m not getting paid for the rewrite since it’s implied that I’m a bad person who broke the contract.”

Duras’ confrontation with AI took place squarely in the arena where the technology is arguably displacing the most human writers: transactional, high-volume content work. But much writing is tied to identity, spirit, and art. This kind of writing would seemingly be the last thing you’d want to give up to a machine. But it turns out, nothing is off limits.

“My master’s students are turning in AI-generated work,” Brian Laidlaw, an author and graduate professor at the University of Denver’s Professional Creative Writing program, told me. Exasperated, he repeated the point: “These are master’s students in creative writing!” I understood what he meant. If you pursue a graduate degree in creative writing, it’s presumably out of love for the craft. 

Laidlaw described his students’ recent writing — the work he suspected was authored by AI — as too perfect grammatically and, the biggest tell for him, humorless. He also pointed out that each student’s work sounds the same: “Genuinely good writing has an aspect of magic to it and serendipity — AI is antithetical to serendipity.” He continued, as if wondering out loud, “Why do it at all?”

In “The Biggest Tell That Something Was Written by AI,” a piece published by The Atlantic in May, journalist Eve Fairbanks offered one possible answer: They do it to stay competitive in fields that continue to value writers who can churn out quality content at high volume. She notes that even professional writers who are confident in their abilities may feel pressure to use AI to gain an edge in their field, whether it’s journalism, academia, or even YouTube influencing.

When I finally took my first tentative sip of the AI Kool-Aid, I saw how easy it is to give in to the machines. It wasn’t that I wanted to avoid writing, but it is damn hard, and after so much rejection, a professional edge was starting to sound really nice. 

AI paralysis sets in

I have one story among my small collection of abandoned fiction that I’ve long considered a lost opportunity. I know it has potential, but despite extensive editing, I’ve never managed to effectively express myself in it. So I pasted it into ChatGPT. Let’s see what AI could do with a piece I couldn’t fully realize. 

First, I asked for general editorial feedback. The evaluation that followed made it sound as though I had written the next great American novel. My “quiet imagery” and “evocative narration” were apparently qualities that “very few people can nail,” the kind of writing you “might see in The New Yorker.”

Knowing ChatGPT’s tendency toward flattery did little to stop me from momentarily considering myself quite the literary hotshot, destined for greatness. Maybe I had captured what I was trying to say after all. After hauling myself out of this self-adulatory pit, I gave more specific prompts: I thought the pacing dragged in the middle, the dialogue sucked, and, overall, the piece read like a series of choppy scenes, not a cohesive whole.

More shallow suggestions followed. I recall squinting at the screen, as if the text were blurry and I needed to strain to read it. Chat (by this point, we were on a first-name basis) sterilized my writing, favoring short, declarative text whenever possible. And don’t even get me started on the dialogue: A scene between two friends read like a Disney movie on steroids, where the characters use language reserved for impossibly naïve people. 

I kept refining my prompts. Eventually, Chat suggested a structural fix that made me pause. “That’s actually not a bad idea,” I thought. At the end of its explanation, it benignly offered up its services: “I can help you incorporate these changes if you’d like.”

I hesitated. For a few long, beautiful seconds, I was still in charge. Then, with frightening speed, I decided that being in charge was overrated. I allowed Chat to draft a few lines, rationalizing that the direction it landed on came from my own steering. Surely I would have eventually arrived at the same place on my own, right? 

I took its text and performed what I told myself was due diligence, copying the lines into a fresh document and using them as a springboard for ideas. By the time I was finished, none of Chat’s original writing remained.

The next morning, I reread the passages. I was squinting again, but this time at my own finished paragraph. I had the sensation that I was looking at something underwater — everything was shimmering and intact on the surface, but there was a thin membrane between me and the words, between me and the ideas they represented, that hadn’t existed before. 

I started to call this sensation — a feeling of disconnection between myself and “my” writing — “AI paralysis.” It comes from moving too easily in a direction you think you’ve chosen but haven’t. I consider it a side effect of what Eric So, a professor at the MIT Sloan School of Management, calls “AI gravity” — the pull of a tool that conserves mental energy, promises an edge, and is hard to resist once everyone around you seems to be using it too. Gravity, in the physical sense, is the force that keeps you grounded. It creates friction. AI gravity removes friction — a relief that is exactly what makes it so easy to adopt.

“It wasn’t writer’s block — I know how that feels,” Sam Andrews, a science communicator, told me. As an experiment, she fed AI fabricated story briefs of varying complexity, along with instructions about word count and intent. She read AI’s take, revised it, reread it, grew frustrated, and started from scratch. “I couldn’t find me in it,” she said.

It was as if her belief in her writing abilities, built and reinforced over years, began to collapse in the span of a few weeks. “I had a huge dip in confidence that was immediately paralyzing,” she said. “These weren’t complicated pieces. But there was this self-doubt in the back of my mind that made me feel like I had to check whether my plan was OK.”

Girl. I feel you. I didn’t need a neurological study (though they do exist) to tell me AI was doing something to my cognition and that it wasn’t good.

The work before the work

AI proponents may point to me as the problematic half in this partnership, arguing that I should treat the tool as a collaborator rather than a replacement. After all, the promise of AI is that it will take on tedious tasks, freeing humans to focus on deeper thinking. Used well, it can theoretically challenge our assumptions, strengthen our work, and help us reach better solutions, faster. 

Jasmine Sun, a prominent Substack writer who covers AI, has encouraged her readers to “be real” with themselves about how they are using AI: “If you pay attention and avoid self-deception, you can feel when you are doing real work.” 

It’s a nice idea. Applied to writing, though, it reminds me of my well-intentioned but perhaps somewhat naïve parents suggesting that I combat my tendency to stress out by “trying not to stress out.” Thanks, Mom and Dad, I hadn’t thought of that. Sun’s method demands exactly the kind of self-awareness that AI — like any tool designed to appeal to our evolved drive to seek out ease and efficiency — tends to erode.

And, in writing, the line between real work and something that feels like real work is often too thin to see. Sometimes, it disappears entirely. 

With this article, there was plenty I did that never made it to the page. I stared out the window of my friend’s living room, where I was house-sitting while I wrote it. I watched sparrows kamikaze his bird feeder. I paced, throwing a pen up and catching it. I ate an entire bag of baby carrots without noticing because an idea had emerged from some back corner of my mind, and I was afraid that the slightest change to what I was doing would startle it away. I wrote and rewrote paragraphs, deleted them, put them back, deleted them again. More than once, I wanted to quit. 

All of it was tedious. None of it was tedious.

Finding the line

Every writer — every person — will have to draw their own line in the AI sandbox.

One well-established author told me he would never let AI revise his sentences or even copyedit his work, but he was comfortable asking it for broad editorial notes. He told me he is “interested in the rules and potentialities of sentences” and wouldn’t turn that over to anyone. His “default interest” isn’t the subject matter, but the act of writing. 

Others draw the line earlier. A science writer I spoke with avoids AI entirely, not because of concerns about authorship or its psychological effects, but because it isn’t always accurate and struggles to place scientific findings in the current context. At best, it wastes her time, and at worst, it can frame results in misleading ways.

Then there’s my friend, a teacher and amateur poet. He told me he enjoyed using AI because it was the only thing that “cared enough about his poetry” to engage with him in conversation about it.

What was Nazir’s line? After interviewing him for more than an hour, I still don’t know. During our talk, he vehemently denied using AI, but his answers to my questions left me unconvinced. The label “AI user” — in a space where it’s not welcome, like a creative writing contest — is also a bit like a digital scarlet letter, one that may follow him forever, whether he’s deserving or not.

But the real reason I’ll never know is because writing has always been a private process: a person alone with their thoughts, wrestling with questions no one else has to face. Today, that includes whether to use an LLM.

I imagine the answer depends on why someone writes, who they write for, and what pressures they face. If AI has given me anything, it’s the impetus to confront those questions for myself.

No shortcuts to the truth

There’s a Thomas Mann quote I scrawled on the first page of my journal after a bad writing day: “A writer is someone for whom writing is more difficult than it is for other people.” 

My hope was that it would serve as a rallying cry the next time I doubted I could write at all — and I often do. When I initially try to tame my thoughts into the constraints of language, I typically fail. But it’s only then, in the effort of writing things down, in those failed attempts, that I realize how half-formed my understanding actually is — my opinions only partially examined, my emotions completely unexplored.

If I want to put anything that matters into words — whether it’s my findings from five years of graduate school, why a particular scientific discovery matters in today’s context, or a message I want to share with someone I care about — I have to stay in that uncomfortable place for a while, where writing is laborious. I have to catch my breath, look around. 

A few days before my Ph.D. dissertation deadline, I holed myself away in my room, sat at my desk, and opened my laptop, determined to finish that night. By the time evening rolled around, I had a solid document I could submit, but I didn’t feel close to done. I put on my headphones, queued up a John Digweed album, and started again. When I finally shut my laptop, I looked at the time — just past 3:00 a.m. I’d spent the last six hours writing and editing nonstop. The Digweed album had played through at least five times, and I was suddenly aware I’d skipped dinner.

No one will read my dissertation in its entirety, and only a few people will read any of it at all. And the content itself — the subject matter expertise I was conveying — is not what I would point to first as my best or most meaningful work. 

But when I think about writing, I often think about that night, when I fell into something that’s the opposite of AI paralysis: pure momentum and focus. It’s a bit like driving a car in traffic, stopping, starting, inching forward, glancing out the window, distractions constantly pulling at your attention. And then you enter a tunnel. Everything closes around you as you race effortlessly toward that little white light — without actually trying to get there at all. 

AI can provide an off-ramp, a quick exit from the most uncomfortable parts of writing. It can take the friction away — eliminate the moments when I stare out a window, disappear into music, draft pages no one will ever see — but I won’t necessarily arrive at the truth, or my version of it, any faster. 

At worst, I may never get there at all.

This article AI hasn’t taken my writing job — but it is living rent-free in my head is featured on Big Think.