Writing as Therapy?

I'll be upfront, the title is a bit misleading. I don't really write lyrics because it's painful. Therapy can mean a lot of things and when people say this in our space I'm not entirely sure what they mean. Everyone has their own way of approaching this hobby and their own reasons why. As for me? What I know is what happened: somewhere in my early forties, after roughly twenty years of not making anything, I picked up a generative AI tool and something switched on inside me that I genuinely thought was gone.

Since then I've written complex rap lyrics, built cinematic music videos, launched a YouTube channel, learned video editing to a functional level in a matter of weeks, and built this entire website to give myself more places to put things. Not in years - in months. I am learning faster now than I have at any point in my adult life. I am more absorbed, more focused, and more addicted to what I'm making than I have been by anything in two decades. And I don't think this is entirely about me. I see versions of this same thing happening to people across the AI creative communities. I'm a Xennial. But it doesn't matter if your an older millennial or a younger Gen X, most of us gave up on creative outlets sometime around their mid-twenties when life moved in and expression moved out. Something about these tools is just.... pulling us back. Something specific is happening here, and I've been very eager to examine this properly.

So that's what this is. I'm putting on my analytical hat here and will be asking: why? What is the psychology of this? What does the neuroscience say? And given what's coming legally and institutionally for the tools that made this possible... where does it all go? Or maybe how long do we have? 🤔

The science of writing it down

Why Expression Is Medicine

Let's start at the foundation. Long before AI entered the picture, psychologists had documented something important about the act of putting internal experience into language. The foundational work here belongs to James W. Pennebaker, a professor of psychology at the University of Texas at Austin, whose expressive writing research beginning in 1986 established that the act of writing about emotionally significant experiences produces measurable improvements in both psychological and physical health.

Pennebaker's 1997 paper "Writing About Emotional Experiences as a Therapeutic Process" summarized results from approximately twenty studies. Students who wrote about traumas for fifteen minutes a day over four consecutive days visited the student health center at roughly half the rate of control subjects in the following six months. Immune function improved. Depression markers decreased. The effects replicated across labs, populations, and decades.

The mechanism, as Pennebaker theorized it, involves inhibition: actively suppressing thoughts and feelings about unresolved experiences requires cognitive and physiological effort that functions as a chronic stressor. Writing externalizes the experience, metabolizes the emotion, and - critically - imposes a narrative structure on what was previously unstructured distress. A 2005 review in Advances in Psychiatric Treatment described it plainly: those who write about emotional topics "generally have significantly better physical and psychological outcomes compared with those who write about neutral topics."

What makes this relevant here isn't just the therapeutic function of writing in isolation. It's what that research implies about people who, for whatever reason, stopped doing it. The person who once wrote - once played music, once drew, once made things - and then stopped, for twenty-odd years, under the pressure of adult life. What Pennebaker's research suggests is that the creative expression wasn't just a hobby. It was processing. And when it stopped, something that needed processing didn't stop. It just went unprocessed.

The Xennial brain in midlife

You're Not Imagining It - Your Brain Actually Changed

Now let's talk about who specifically is experiencing this creative reawakening - and why the timing matters. I'm a Xennial. Born in that narrow band between 1977 and 1983 that sociologist Dan Woodman at Melbourne University identified as a micro-generation distinct from both Gen X and Millennials. As researchers describe it, we had an analog childhood and a digital early adulthood - the last generation to remember organized childhoods without screens, and the first to adopt every subsequent technology wave as young adults rather than being born into it. We grew up with cassette tapes, played Oregon Trail on early computers, got our first cell phones in our early twenties, and learned to build a social media presence in our thirties.

This dual fluency - the ability to read both analog and digital environments with equal ease - turns out to be psychologically significant when you meet a tool like generative AI. We're not intimidated by it the way older generations sometimes are. We're not naive about it the way younger ones sometimes are. We have enough accumulated life experience to have things to say, and enough technical comfort to pick up the tool quickly. That combination is, it turns out, extremely good conditions for a creative explosion.

And here's the neuroscience that underpins why: midlife brains - in the 40s specifically - are measurably better at pattern recognition, connection-making, and integrative thinking than younger brains. Not at everything; processing speed declines. But the ability to synthesize accumulated knowledge, to draw unexpected parallels, to construct meaning across domains - these peak or near-peak in midlife. The experience of twenty years of adult life, even if no formal creative work was happening, was still loading the mental library. The brain was still making connections. The raw material was accumulating even when the output had stopped.

A landmark 2025 study published in Nature Communications measured what happens to the brain during sustained creative engagement, using brain-age markers to track biological aging. The finding was striking: people who engaged in creative activities - music, visual art, dance, even certain video games - showed measurably younger brain-age profiles. The effect scaled with expertise. And critically, even short-term creative training produced detectable changes in just several weeks. The brain responds to creative engagement with something that looks, at a biological level, like a reversal of the clock.

I don't think this is coincidental. The people I see in AI creative communities who've gone from dormant to obsessively productive aren't just having a hobby. They're responding to something their nervous systems have wanted for a long time, given back to them through a door that wasn't there before.

What the tool actually does

AI Didn't Give You Creativity - It Removed the Barriers to It

This is the distinction I keep coming back to. I don't think generative AI created my creativity. I think it systematically removed the specific obstacles that had been standing between me and it for twenty years. To understand why that matters so much, you need to understand what those obstacles actually are.

Mihaly Csikszentmihalyi, the Hungarian-American psychologist considered the founding figure of "flow" research, spent decades studying what happens when people enter states of complete creative absorption. His formulation of flow - described as "a state in which people are so involved in an activity that nothing else seems to matter; the experience is so enjoyable that people will continue to do it even at great cost, for the sheer sake of doing it" - And this is central to understanding what creativity-unlocking feels like from the inside. It's not effort. It's the opposite of effort. It's being pulled by something intrinsically rewarding rather than pushed by obligation.

Csikszentmihalyi identified the conditions required for flow: the task has to offer a genuine challenge, but one proportionate to your current skill level. Too easy and you're bored. Too hard and you're overwhelmed and quit. The window between boredom and overwhelm is where flow lives - and where creativity happens. The problem with traditional creative tools for someone who hasn't practiced in twenty years is that the technical skill required to produce anything resembling what's in your head is massive. You can hear the music. You can't play it. You can see the video. You can't edit it. The gap between imagination and execution is crushing enough to kill the impulse entirely. And for most people, it does. That's why they stop. And I mean....yeah, duh....i wrote like 50 songs last year...how else do you explain that?

A 2025 study in Frontiers in Computer Science examined what they called the Human-AI Co-Creative Design Process (HAI-CDP) and found that AI tools "lower the entry barrier to creative engagement" in measurably different ways for different people: for novices, the primary value is in breaking through early-stage cognitive barriers that block idea generation. For people with accumulated creative experience and taste - like a Xennial who spent twenty years consuming music and film and art and ideas - the AI accelerates and refines rather than generates. The direction and judgment remain human. The execution friction is removed.

Research published in Science Advances confirmed that access to generative AI ideas causes creative outputs to be evaluated as more creative, better written, and more enjoyable - and crucially, "especially among less creative writers." The people who benefit most from AI as a creative tool are not the virtuoso professionals. They're the dormant creatives. The ones with twenty years of compressed ideas and no viable channel for them. Sound familiar? That's a precise description of folks around their forties who just discovered these tools. 

The obsession explained

Why You Can't Stop

There's a specific quality to this creative reawakening that I want to name directly, because I think it's important to understand it clearly. The compulsive quality - the inability to stop, the way it consumes time and attention at the expense of other things - has a precise psychological explanation. It's not weakness or lack of discipline. It's dopamine architecture meeting intrinsic motivation, and they're both firing at once.

Csikszentmihalyi described the flow state as "autotelic" (I had to Google that) - it means the activity is its own reward. The experience of making something is intrinsically satisfying independent of outcome, audience, or recognition. His research showed that flow experiences produce what looks, neurologically, like transient hypofrontality (😅) - a temporary reduction in the prefrontal cortex's self-monitoring activity that removes the inner critic, produces a loss of self-consciousness, and generates the subjective experience of being "carried by a current." Its like time itself distorts. External concerns fall away. The person is completely present in the act of just ..."making".

For someone who hasn't experienced this in decades - who has been living, in Pennebaker's terms, with a persistent inhibitory cost - the first time the flow state returns through a new creative tool, it doesn't just feel good. It feels like something that was missing is back. The addiction quality isn't to the tool. It's to the state the tool produces. And that state is not a luxury. The research suggests it's a functional necessity for wellbeing that a significant portion of adults have been without for a very long time.

The website. The YouTube channel. The song after song after song. These aren't symptoms of something going wrong. They're symptoms of something that was always there, finally getting out. The question is whether the obsession can be directed rather than just experienced. Whether the creative energy that AI has unlocked can be channeled into something durable, or whether it will eventually exhaust itself on the treadmill of output for its own sake. I don't know the answer to that yet. But I know the question matters, since I already can see folks in the community burning themselves out by the looks of it.

The legal ground beneath us

The Industry That Sued Its Own Audience

Hard pivot now... and a bit of context we need.

In June 2024, the Recording Industry Association of America - on behalf of Sony Music Entertainment, Universal Music Group, and Warner Records - filed landmark copyright infringement lawsuits against both Suno and Udio, alleging both platforms trained on copyrighted recordings at massive scale without permission or compensation. Suno responded arguing fair use - that its model learned from music the way a human musician does, by listening, not by copying. Then, fairly quickly, the industry's fighting posture started to look like something else entirely.

You're welcome to read WMG's "victory for the creative community" framing however you like. What it actually describes is a major label that sued an AI company and then signed a partnership deal with the same company once the legal leverage produced favorable commercial terms. The artists whose recordings allegedly trained the model without consent are not mentioned in the settlement terms, because that's not how these things work and never has been.

Which brings me to something I feel strongly about, and want to say clearly: I don't have much patience for the music industry or some tech bro positioning itself as the moral guardian of creativity here. Not because the copyright arguments are necessarily wrong in a legal sense - but because the industry bringing them has its own history to answer for. Spotify's streaming ecosystem has been documented as riddled with fake-artist fraud, bot networks, and what investigators uncovered as an internal Spotify program - called Perfect Fit Content - where cheap tracks were commissioned under fabricated artist names and placed on curated playlists, paying lower royalties than real artists would receive. Rolling Stone's investigation into streaming fraud found the practice so embedded that artists, managers, and labels were all participating - inflating streams through bot farms to game chart positions and royalty calculations. The streaming revenue that musicians now say is being cannibalized by AI was already being cannibalized systematically, structurally, from the inside, long before the first Suno User released his first track.

I understand where working musicians are coming from. The anxiety is real and the displacement concern is legitimate. But the labels standing in court framing this as a fight for artists are the same entities that built a streaming infrastructure where a significant portion of the plays were never real to begin with. That context belongs in this conversation.

Where this leads

The Genie Doesn't Go Back in the Bottle

Here's where I stop pretending this is a debate with an open outcome. The Western legal system is having a very serious conversation about AI and copyright, in very serious courtrooms, with very serious lawyers - and none of it matters in the way the people having it seem to think it does. Not because the law is irrelevant, but because the law is local and the technology is not.

While the RIAA litigates in Massachusetts federal court, Chinese AI music models are training on everything, bound by no Western intellectual property framework, subject to no American court ruling, and getting better at a pace that makes the Suno debate look like a dispute about candle manufacturing the year the lightbulb was invented. Udio can settle. Suno can license. A ruling can come down tomorrow that reshapes the entire US AI music industry - and a model released out of Shenzhen, trained on the latest Spotify leak, next quarter doesn't care. Neither does the open source ecosystem, which has already produced locally runnable music generation tools that no courtroom on earth can decommission. You can win every motion in that room and still lose to some kid with a gaming PC and a download link. It's a cease-and-desist served on the weather, and the weather does not care. These are two separate conversations that only one side seems to realize aren't connected.

And then there's the question nobody in the legal debate wants to answer directly: what happens when it becomes genuinely indistinguishable? Not almost. Not close. When a trained ear in a blind test cannot tell the difference between an AI-generated track and one made by a human producer. We are not far from that point - some would argue certain genres are already there. At that moment, the entire framing of the debate collapses. You cannot protect something you cannot identify. You cannot gatekeep authenticity when authenticity is no longer audible. The industry's argument depends, at its foundation, on there being a detectable difference worth protecting. That foundation has an expiration date, and it's approaching faster than any appellate court moves.

What this means practically - for me, for the communities I'm part of, for every dormant creative who found their way back through these tools - is that the creative reawakening doesn't have an off switch. Even in a world where every Western AI music platform gets litigated into licensing compliance or shutdown, the tools exist, they're spreading, they're getting better, and they're increasingly running on hardware you own. The genie left the bottle the moment the first model went open source. Everything since then has been legal theater performed for an audience that doesn't quite realize the stage is already empty.

For me personally - and I suspect for most similar aged folks in this space - what matters most was never the platform. It was what the platform proved: that the gap between having something to say and being able to say it could be closed. That twenty years of quiet didn't mean the voice was gone. The rap lyrics about borders and fear. The website built because there were things that needed saying. The compulsive making of worlds out of analog memory and digital tools. That didn't come from Suno. It came from somewhere that no court ruling touches.

Pennebaker would recognize it. Csikszentmihalyi would recognize it. The tool was never the point. The expression was the point. It always was.

Cheers,
Aidan