How I Use AI and How I Don’t Use AI

Tech July 14, 2026 17 min read

How I don’t use AI

I don’t use generative AI to write prose or lyrics. I don’t use it to write blog posts. I don’t use it to write e-mails for me, outside of the occasional auto-complete of what I was going to say anyway. I don’t use it create music. I don’t use it to create graphics. I consider those lanes of creativity to either be second nature already or rewarding forms of toil — like a zen garden or a Bonsai tree. I happily do them the old fashioned way, with no generative AI but some minor augmentation via machine learning through plugins, which is essentially inescapable in the modern digital landscape.

You’ll notice that I used an em dash in the last paragraph even though I claimed to not use generative AI to write prose. That’s because I’ve been using em dashes for decades. I find them to be more elegant than semicolons in nearly every way. I also tend to hyphenate small groups of words to enhance their readability, which is apparently another thing that AI does a lot. Sorry for teaching it stuff. I’m not going to change how I’ve been writing for my entire life to pass an AI purity test though, so we’ll just have to use the honor system if that’s alright.

How I Use AI

I use AI when faced with annoying toil, which mostly ends up being the iterative cycle of trial & error that is inherent in software development, or boring documentation. AI makes my documentation much more robust than it would be otherwise because I am lazy. I have a knack for writing good documentation thanks to several years of working in support, but I don’t like doing it by any means. So I tend to have AI write it and then I heavily edit it down to something usable.

I have never had any particular talent for — or interest in — writing beautiful code. I have always found coding to be kind of annoying process. I have diagnosed ADHD and suspected AuDHD, so I tend to either hyper-focus on things or not focus at all, which is a bad trait to have as a programmer because I will either spill 2 days of my life out on a single, potentially small problem that I’m trying to solve programmatically, or I forget about what I’m doing all-together and drag my feet to get back to it. Sometimes I never get back to it at all, resulting in a massive graveyard of dead projects that is a familiar fixture for most software developers.

When I code, I’m typically just trying to find the quickest path to solve a problem, automate something, or to make something easier for myself to do again in the future. I’m very lazy and inefficiency makes me mad — literally, it does. That’s why I’ve made a career out of solving problems and automating stuff. Things that get in the way of my ability to be lazy, like manual tasks with high probability for human error — or strings of things I have to remember to do in sequence — always feel like an open sore on the roof of my mouth that I can’t leave alone once I know it’s there.

I am an autodidact, which means that I am self-motivated to learn through self-study, and so I teach myself stuff all the time but combine that with my ADHD and it means I have major trouble focusing on subjects that I’m less interested in. Luckily, I’m in love with computers, so I poured my entire life into learning about them — and I still love and learn. I’m usually somewhere near the cutting edge of the goings-on in tech.

Unfortunately, it wrecked my formal education and I never saw the 9th grade. I naturally seek to know a good deal about a lot of stuff, and I can hyper focus on subjects I’m interested in and learn a lot about them in a small amount of time, but I don’t generally have a master’s level of knowledge about any one thing in particular. I’m a Jack of All Trades type of guy and always have been. There have always been weird gaps in my knowledge that I have to backfill as needed — if needed.

Prior to the introduction of large language models, most of my projects were ugly amalgamations of code blocks that I found on the Internet somewhere and mashed together until the thing I was coding did what I wanted it to do. It took me days or weeks to come up with something usable most of the time. I actually hated that process, but hated not solving a problem even more, so I typically endured the pain — especially if the problem was someone else’s.

The LLM essentially generates code in the same way that I would have done myself, but much faster, with fewer syntactic blunders, and zero simple errors in logic. This new form of iteration feels like rewarding toil to me, like writing music, versus the annoying toil of blind trial & error — or Googling in a desperate attempt to find someone, anyone, who had the same problem in the past and remembered to post a solution to some random forum.

It’s taking my prompt and going to the same places I would have gone, but parses the output way faster than I ever could. I prompt it based on pretty good fore-knowledge of what I’m doing and it backfills the gaps in my knowledge where needed, but my experience as a professional nerd allows me to notice when the AI is doing something in a dumb way or hallucinating, so I can quickly stop it and correct it. The result is a workflow that enhances my productivity while also no longer feeling like work.

I have been coding with LLMs since they came out and I’ve gotten pretty good at writing efficient prompts, knowing which tools are going to be a waste of time and which are going to bolster my workflow, and getting good code out of an agent running the frontier models. The entire time I’ve used LLMs, I’ve never once felt like my livelihood was being threatened by them. They can’t reason like I can, they can’t detect shitty UX, or act upon hunches built on the context of an entire career in several areas of tech and related fields.

Not only will a LLM never replace me, they make me feel more required — more secure. Putting a LLM in the hands of a novice engineer is like putting a Cessna pilot in a F35 cockpit. They may be able to fly, maybe, but they can’t get anywhere close to the performance of a trained F35 pilot. I’m not a F35 pilot but I’m at least a F16 pilot in this metaphor. I’m in the ballpark.

My Other Thoughts On AI

I am very optimistic that LLMs will lead to many advancements that will have a positive impact on humanity. I see the potential already, in this incredibly young technology, and I know what happens historically when smart people get their hands on new and amazing tools. Society always benefits. I think we’ll come out on top of this one.

I also harbor kind of a radical ideology as it relates to intellectual property, in that I have always thought that information should be open and free for everybody to use. I grew up poor and free information enabled me to learn whatever I wanted to learn, which is now powering my entire career as an adult. Without the Internet, I believe that I may have never known any success in my life.

When I think about large language models, I have see a portable, compressed version of the Internet that lives inside of a big indexed spreadsheet, easily searched, easily utilized — and I believe that all of the information therein should be free in the first place. The LLM represents another layer on top of the Internet, and it’s a good one.

When you ask an LLM a question through a chatbot, it generates a sequence of words with their plausibility determined by applying statistical patterns learned from vast amounts of human writing, with ZERO grounding in truth, experience, or intent. Whatever intelligence they have, it isn’t human-like understanding. The conversational persona is an interface design choice — a mirage of marketing. There is no market in a LLM’s ability to impersonate a human outside of bad companies who are okay with bad customer experience. I think we’re all discovering that in real time.

I think there is nuance in the debate about whether or not generative AI can be art or not. I think right now, it’s not art, but it can be the catalyst for art. It reminds me of a couple of styles of art that were born out of theft and crime in the 70’s and 80’s — graffiti and sampling.

Graffiti started out as amorphous blobs of paint being shoved into everyone’s faces on vandalized trains, simple tagging by bored teenagers, gang members, or whomever — people that had time, paint, and not much to lose. Over time, an industry popped up around the style, leading to custom graffiti tools — like caps — that made it easier to make stuff look cool if you took the time to become skillful with the new tools. Now your city probably has a beautiful graffiti mural somewhere in it, because the art had time to evolve with the artists.

I am actually looking forward to this kind of a revolution in filmmaking because good film is incredibly cost prohibitive. Getting into film making is a huge gamble that almost nobody wins. The industry built around it is for the benefit of the producers, the people who already have the money, not the artists, and they certainly don’t let subversive content through most of the time. I’m waiting for a hip hop style revolution to happen in film because there is a need for subversive messaging, now more than ever. I want young, angry kids, hungry for success, to have access to tools that allow them to tell the exactly story that they want to tell about the injustices they’re facing. I want them to be able to make ultra high quality films without having to go through someone who’s entire goal is to oppress them. I truly believe that’s what generative AI is going to provide in film.

I’ve been a fan of hip hop since the 1980’s and at several points in my life, I’ve considered it my favorite genre of music. It’s built on sampling, which is when someone with a couple of turntables finds parts of a record that they like, records them, mixes samples from the records, and adds other production elements to make a new song.

There was a famous court case in 1991 (Grand Upright Music v. Warner Bros.) that saw the rapper Biz Markie successfully sued by Gilbert O’Sullivan for sampling the song “Alone Again”. As a result of that decision, any samples now have to approved by the copyright owner, or all of the money made by the new song goes to the original copyright owner — usually a record label. That new precedent created a very lucrative revenue stream out of thin air for the major labels right as hip hop was entering into the main stream.

I do believe that artists should own the rights to their work, and I don’t like the rapacious methods that the AI companies utilize to train their models, such as buying books, scanning them, and destroying them. If they are obtaining copyrighted material to resell as their own product practically verbatim, I don’t like that. I don’t like that Spotify likely began with a largely pirated music library and sold streaming access to that. I definitely don’t like the AI companies stole Spotify’s music, audiobooks, and podcast libraries to train models for the purposes of generative AI to enable the mass production of soulless art. I want kids without financial means to have access to tools to make high quality art that they couldn’t have made otherwise. I don’t want companies using it to flood the Internet with single prompt slop songs that suck. The former feels righteous. The latter feels icky.

I think that the actual crime is in how the models are trained, and not the output of the model, if an artist is actually shaping it into something new and novel — something good. I’m not talking about a few prompts to make something that looks like a Studio Ghibli character. I mean something like a piece of art that stands on its own and its AI origins are a distant afterthought. I don’t know if we’ve had that kind of art yet, but I think it’s coming because I believe in the creative power of humans. Someone is eventually going to invent a novel art style that uses generative AI, and I would look at that with the same rose-colored glasses that I use to look at sampling in hip hop.

I believe that sampling is an incredibly valid art form and I always believed that 1991 court case against Biz Markie was bullshit. Every artist that I have ever known believes that at its very core, art is theft. Artists constantly steal from other artists and bite each-other’s styles. That’s why eras of styles exist like Renaissance, Baroque, Romanticism, Realism, Impressionism, and on and on. There are many, many artists who produced art in those eras that you can identify at a glance because they were all learning from one another — what they were stealing from each-other. The same thing goes for music. Every art form is packed with artists stealing from other artists all the way down. I see multiple songs I like every day on Instagram Reels from different girls sitting cross-legged in their bedrooms with a MIDI controller, looping and singing in the same genres. I’m not mad at it. I love art.

The most recent famously egregious sampling crimes I can think of off the top of my head is Lucid Dreams by Juice WRLD, which utilizes a barely altered sample Shape of My Heart by Sting. Juice WRLD (RIP) was an incredibly talented lyricist, took the guitar riff and slightly lowered the pitch and speed, and turned it into a new and beautiful thing. Even Sting thought so, but there were also rumors that Sting received 85% of the revenue generated by the song and I hate that.

I hate that, and most other sampling revenue splits, because I know the history of hip hop, sampling, and where they came from. To get to the perfect 10/10 album (in my opinion) To Pimp A Butterfly by Kendrick Lamar, there first needed to be a bunch of poor kids in the middle of a the 1977 blackout of the Bronx, looting a bunch of electronics stores and stereo equipment stores. They stole turntables and mixers that they could have never afforded otherwise. Suddenly, teenagers without the means to have one turntable suddenly had multiple turntables, and they learned how to use the mixers. They used stolen records from stores, or borrowed records from their parents, and they played around until they invented a new art form.

It’s actually my favorite “fuck you” of all time, in fact, if you look back far enough into the past. I’m going to write out a brief history of hip hop to highlight my point.

Enslaved Black people, working in fields, sang spirituals with coded, double-meaning lyrics — songs that could carry hidden messages or mock slaveholders who often clapped along, oblivious.
Spirituals, work songs, and field hollers evolved into the blues (roughly late 1800s–early 1900s).
Newly freed black communities built the blues and juke-joint scenes.
White audiences were increasingly drawn into black music venues and markets.
The white-owned record industry began commercializing the blues, frequently underpaying black artists and leaving them uncredited for their music and/or signed to exploitative contracts.
The blues fed into rhythm & blues (R&B), a term coined in the late 1940s to replace the industry’s “race records” label.
Rock and roll grew directly out of R&B in the early-to-mid 1950s, led by black artists like Chuck Berry, Little Richard, and Fats Domino. Then white artists like Elvis covered and popularized the sound for mass (white) audiences.
The white covers of black music often outsold the black originators, because they had the means to record and distribute at higher quality and volume, with songwriting credits and royalties frequently going anywhere but to a black musician or songwriter.
The Civil Rights era ended legal segregation (mid-1960s).
Institutional oppression was ongoing, like redlining, disinvestment, and the concentration of black and latino communities in neglected urban areas like the South Bronx in New York City.
Hip hop emerged in the Bronx in the early 1970s with foundational moments like DJ Kool Herc’s 1973 block party, then Grandmaster Flash and Afrika Bambaataa soon after.
In July of 1977, a massive blackout hits New York City, and the Bronx particularly hard. Kids with no money, no promise of a future, and nothing better to do go on a looting spree, emptying electronics and stereo equipment stores. They stole turnables, mixers, and records en masse. This accelerated an already-growing block party scene.
DJs began experimenting with two turntables (when they couldn’t afford one) and breakbeats. They invented turntablism (scratching, cutting, looping the break). Hip hop was growing and growing fast.
Many crews formed with DJs utilizing the stolen turntables and MCs began to rhyme over the beats that were built using samples and loops of music — most of it recorded based on styles of music that had been stolen from their ancestors and commercialized by the labels that were now scrambling to throw money at these new artists.
The pillars of hip hop culture begin taking shape and breaking into the mainstream: DJing, MCing, breaking, and graffiti. I remember laughing at the white people on TV doing it all very badly in the early 80’s, as seen in Rodney Rappin’ and Breakin’.
Some artists and entrepreneurs begin to achieve real commercial success. Labels are chasing and signing all of the rappers that they can find.
Grand Upright Music v. Warner Bros. (1991, the Biz Markie case) establishes that samples must be cleared, so labels owned by rich white elites (mostly) can not only demand a new stream of profits, but also stomp on competition, or use the legal precedent as leverage in negotiations, once again stealing the profits from a new art form pioneered by the black community.
Sample clearance becomes expensive and legally risky, which ends the era of dense, freewheeling sampling, and reshapes hip hop’s economics to make it impractical for artists without means to break in to the industry. Time and time again, young new artists blow up with a popular song that has an uncleared sample in it, and they’re forced to redirect all of the profits to an already rich record label.
Generative AI models are trained on popular music, most of it owned by major labels. Those same labels already pay artists next to nothing: standard contracts turn them into unrecouped debtors stuck in deals they can’t exit, and “360 deals” now claim cuts of their touring and merch income too. Meanwhile, the labels negotiated sweetheart deals with the streaming services, which pay independent artists next to nothing — fractions of a cent per play.
Everybody seems to be deeply offended by generative AI risking the intellectual property of the major labels, publishers, and film studios, which are a very large portion of the training data used to train the models.

Having the knowledge of what needed to happen to get to Kendrick Lamar — the righteous theft and ensuing creativity that gave birth to a couple of my favorite art forms — leaves me a little less concerned with the impact of generative AI on the arts. I trust humans to make cool new shit when they have the tools at their disposal to do so, and I think that generative AI might make things a lot cheaper than they would be otherwise.

That said, I’m not fully sold on it yet. Local inference, or using your own personal computer with a local LLM, is how this becomes an effective form of art for the purposes of protest. There is progress in that direction but it’s still a-ways-away. So my big caveat is that I think generative AI will actually enhance the ability of artists stuck in poverty as long as it doesn’t require high-cost subscriptions. I think as the technology evolves, that’s not only very possible but the likely outcome. I think that the data center heavy iteration of the LLM boom is like the IBM mainframe era. Relatively short-lived and necessary, but usurped by on-prem solutions as soon as it was financially viable.

The power consumption, the water usage — the environmental impacts — are not lost on me and are certainly not to be ignored. I don’t like how the billionaires are rushing the infrastructure projects in the same rapacious fashion that robber barons usually do. I think it will push the technology forward rapidly, because that’s one of the things capitalism does well, but I still don’t like how they’re doing it.

I do still believe that AI will have a net-positive impact on society at large, having seen first-hand how quickly it has already evolved since being introduced. The impact on STEM research is going to be enormous. I’m optimistic that bolstering our smartest researchers with large armies of polymath PhD level AI agents can only be a good thing for our prospects of reversing catastrophic climate change patterns. Through decades of inaction by our leaders, that I have had to frustratingly live through, we’re probably too late to remediate the problem through the traditional means of conservation. The Atlantic Meridional Overturning Circulation is provably weakening already. Once it tips toward collapse, things will begin to change until the Earth becomes unlivable for a large number of its current inhabitants.

I am of the mind that one of our only chances to avert the disastrous climate consequences of our unfettered industrialism is through some eureka-style innovation, and I just think our chances of doing that in time are much higher with the assistance of evolved forms of AI. That evolution of a new technology is going to have to go through a period of inefficiency, like every other new technology, to get to its most useful state. That period of inefficiency, while we innovate, is going to feel incredibly counterintuitive but I think it will be worth it. I don’t think it’s worth shoving data centers into neighborhoods that aren’t zoned properly for them, but I think it’s worth deploying the the technologies intelligently and seeing where we can take it.