Learning Music with AI

How Being Bad at Something Made Me Want to Try Again

When Curiosity Meets Creative Resistance

There’s a strange moment that happens when curiosity bumps into resistance.

You know the feeling. Someone shares a thing. You look at it. You instinctively recoil. Not because you hate it, but because it challenges something you already believe.

That was me the day someone dropped a link to a song they made using AI.

My first reaction was simple: What is this?

Once I realized it was AI-generated music, I was already halfway out the door mentally. But if you’ve been hanging around this corner of the internet with me for a while, you know I have a habit of poking at new tools instead of ignoring them. Not out of blind enthusiasm — mostly out of curiosity. Writers, artists, and creatives don’t get to pretend technological shifts aren’t happening. Even if we choose not to use them, we still need to understand them.

So I listened. It wasn’t bad. It wasn’t great. Something about it felt off. Not wrong exactly — just missing something human that I couldn’t articulate. The problem was that I didn’t have the vocabulary to explain why and I felt like saying ‘AI made’ wasn’t a strong enough justification.

The main issue is that music has never been my lane.

Why I Never Learned Music…

I played the recorder in elementary school like every other kid forced through the rite of passage, then spent two awkward years in band loudly contributing trumpet sounds that could generously be described as enthusiastic. If you want to know my ranking, it was second to last chair. One time I got to second chair to spite someone and then went back to second. My instructor wasn’t impressed. After that, music became something I consumed, not something I made or help create. I sing along in the car. I hum while working. But actual music creation? That felt like a language I never learned.

Maybe part of why music never stuck was that I never gave myself permission to be bad at it long enough to improve.

And yet, that curiosity lingered.

I have a guitar sitting in my office. It’s been there long enough to feel more like decor than an instrument. Every now and then I look at it and think, I should learn to play. Then life happens. Writing deadlines happen. My boss decides today is the day to release a new book. Business tasks happen. The guitar sits forgotten in my craft closet.

Trying AI Music for the First Time

So when I saw the AI music tool offering a free trial, curiosity won — I signed up.

Naturally, my first test subject was “Twinkle, Twinkle Little Star.” It’s the only song where I could confidently recall both lyrics and melody without Googling anything. Not exactly ambitious, but it gave me a baseline. I wasn’t trying to create art. I was trying to understand the machine.

The more I experimented, the more I realized that the tool itself wasn’t the interesting part. The learning curve around it was. To get anything remotely good out of the program, I had to start learning about music in ways I never had before — lyric structure, genre expectations, rhythm, syllable stress, phrasing, and how tone shifts depending on style.

The Unexpected Lesson: Permission to Be Bad at Something

But more than that, I discovered something I didn’t expect to enjoy: the freedom to be bad at something in private.

When you’re learning a new creative skill publicly (in classes or with an instructor), there’s often an invisible pressure to perform competence before you actually have it. You hesitate to experiment because someone might see the messy middle, watch you fumble, or even laugh at your stumble. You avoid trying because the early attempts rarely reflect the vision in your head.

With my experimenting, I could try something, listen back, cringe a little, tweak it, and try again — all without anyone watching. It was like rediscovering the scientific method we all used instinctively as kids. Hypothesis: What happens if I change this line? Experiment: rewrite it. Observation: that didn’t land. Conclusion: try again with different rhythm.

Repeat.

Over and over.

How Experimentation Turned Failure into Information

I remember one lyric attempt where I tried to force a dramatic phrase into a light, upbeat melody. On paper it felt clever. In playback it sounded like the musical equivalent of tripping over your own shoelaces. I laughed, rewrote it, and tried again. The next version wasn’t perfect — but it was better. Not because I suddenly understood music, but because I had collected one tiny piece of data about what didn’t work.

Failure stopped feeling like failure and it became information for me to process, reflect, and learn from. Somewhere in that repetition, the thrill crept in. It encouraged me keep trying when I would have been frustrated and walked off. This was new and exciting for me and I found motivation where I had none before.

Not the thrill of producing something — that came rarely — but the understanding just a tiny bit more than I did before. Each awkward lyric taught me something about cadence. Each genre experiment showed me how structure shapes emotion. Each “off” result sharpened my ear in ways passive listening never had. That shift was addictive in the best way.

What AI Music Taught Me About Creativity

I remembered working with the Laylines Band on Down Way Out. Caleb patiently explained why certain words landed well in rap sections while others felt clunky. I nodded along at the time, convinced I understood. Looking back now, I realize I grasped the concept intellectually but not experientially. There’s a difference between being told why something works and wrestling with it yourself until it clicks.

I spent weeks crash-coursing my way through musical concepts — not in a formal, academic way, but in the messy, trial-and-error way creatives tend to learn when mommy and daddy aren’t paying the bill. I rewrote lyrics repeatedly. I experimented with genre shifts. I listened critically instead of passively. Somewhere between chasing rhyme schemes and testing tempo changes, I started glancing at the guitar in my office differently — less like decoration, more like unfinished business.

Can You Tell the Difference Between AI Music and Human Music?

And here’s the funny part: seasoned musicians can still hear the AI in the music.

There’s a polish that feels slightly too clean. A structure that occasionally leans predictable. A performance quality that lacks the micro-imperfections humans naturally bring. But the average listener? They often can’t tell the difference. That realization sits somewhere between fascinating and unsettling.

Because, as I like to remind everyone, this technology will get better. It always does.

AI as a Creative Sandbox, Not a Replacement

The conversation around AI in creative spaces is often framed in extremes — either it’s the future savior of creativity or the destroyer of human artistry. My experience landed somewhere far more nuanced. The tool didn’t replace creativity. It exposed my lack of musical literacy and gave me a low-stakes environment to explore it.

AI didn’t make me a musician. It made me curious about music in a way I hadn’t been before. It turned passive appreciation into active experimentation. It pushed me to think about lyrics as more than poetic lines — to consider timing, breath, phrasing, and how words sit inside melody.

Why This Experiment Changed How I Look at My Guitar

And unexpectedly, it changed how I look at that guitar in my office.

Before, it represented a vague aspiration. Something I thought would be cool to learn someday. Now it feels like a continuation of an experiment already in progress. I’m not starting from zero curiosity anymore. I’m starting from momentum and motivation and that can be a better starting block than just the decision to try something new.

The Joy of Learning Something Messy and New

I’m still cautious about AI. I still have questions about ethics, authorship, and long-term creative impact. Those questions aren’t going away anytime soon. But I can also acknowledge that, in this small corner of my creative life, AI acted as a doorway rather than a replacement.

It lowered the barrier to entry, gave me permission to experiment without performance anxiety an reminded me that creative growth is often messy, repetitive, and quietly joyful.

I don’t see myself signing up for open mic nights any time soon. But I do see myself scribbling lyrics with melody in mind. Fumbling my way through chord charts. Picking up that guitar in my craft closet not to perform, but to play and see what happens. To fail and fiddle with something privately and try again until it feels right.

Maybe we all need a private sandbox again — a place where curiosity matters more than competence.

Creativity Starts with Curiosity, Not Competence

Sometimes creative growth doesn’t arrive through the doors we carefully plan. It slips in sideways — through a tool we doubted, a song we almost skipped, or a free trial we expected to forget. What began as curiosity quietly became momentum. And in that momentum, I was reminded of something simple but easy to forget: being bad at something isn’t a failure. It’s the beginning of fluency. So for now, I’ll keep experimenting — awkward lyrics, strange genre mashups, and tentative guitar chords included — trusting that curiosity is often the first note of something worth learning.

Thanks for reading my fun Friday post. I’ll see you next week.