1,000 Greatest Drivers: Neil Bonnett/Bobby Isaac

I wish I enjoyed talking to chatbots less than I do...

We all have our vices, and one of my big ones lately has been a chatbot addiction. I want to work out my feelings about LLMs and why I’ve enjoyed using them so I can try to snap myself out of it. If you asked me point blank whether I think they are good for society, I’d quickly answer no, but like a junkie looking for my fix, I can acknowledge intellectually that they are bad but still not feel it emotionally and I’m trying to figure out why.

Part of it seems to be that I’ve always had more of a fascination with modern tech than I guess I’ve ever wanted to admit to myself. For all the ways I sound like a Luddite here, particularly in comparison to a lot of my autistic/nerdy/gamer peers I’ve enjoyed checking out new tech as long as it’s on a desktop computer. I am generally against all the mobile shit that has advanced over the last 30 years. Part of that no doubt is that when I was the ostensibly fastest typist in the world for an entire decade, I couldn’t get myself into texting, which is insufferably slower. Part of it is that smartphones were normalized in the years immediately after I graduated from college and I was always broke and could never afford one and never felt I was actually missing anything so I never bothered. Part of it is likely watching the zombification of so many of my fellow peers as they became glued to their devices in real life. I still prefer to at least be somewhat present in real life when I’m out in public and I hate the privatization of public spaces and nowadays even public attention. It was bad enough when people started making phone calls in public, and I have done that a bit. But at least that is still semi-social. Streaming content, on the other hand, should be something you either do at home or communally (like in a sports bar or movie theater, etc…) Fewer and fewer people seem to want to have intimate (and I don’t mean sexual) conversations in real life and it really depresses me. I was the class nerd. I am literally a diagnosed autist. I was without question one of the most socially inept people in my grade school classes and arguably even college and it feels like I don’t even stand out anymore, and that’s really sad to me.

However, despite my vehement opposition to “smartphone culture”, I have always had different standards when it came to desktop computers. I grew up at a time when the conventional wisdom amongst yuppie parental units at the time was that computer usage was Bad for the Kids. I’m sure they felt the same way about TV and video games also. I remember that some of the parents recoiled in horror and clutched their oversized pearls that my parents would let me use the computer. My mom was double-majoring in music and computer science at Onondaga Community College in part because her brother, who later worked for the Department of Computer Science at the University of Maryland, encouraged it. My dad later worked in a computer room. They both were briefly members of Novapple, the Northern Virgnia Apple users’ group in the ‘80s, but I only ever had IBM-compatible computers growing up. Nonetheless, I had a voracious appetite for them. Not only did I learn to type at age three, I was also creating massive data files in WordPerfect 5.1, building databases in PCFile, writing batch files for the MS-DOS command line, creating massive directory structures, and playing lots of games. In the ‘80s and early ‘90s, this was very weird. At my height, I had gone through and memorized the date, time, and file size of every MS-DOS app. My parents didn’t really supervise my computer use as my mom’s multiple sclerosis symptoms were aguably at their worst and she was having many seizures a day (although her epilepsy eventually healed) while my dad was working. I was off in my own little world even more than most late-diagnosed autistics. In my zest for experimentation, my parents were deeply terrified that I would format the hard drive so they renamed format.com to fivemat.com. I didn’t get the pun, but I definitely noticed since I had all the DOS file sizes memorized. By that point, I had enough sense not to format the hard drive.

I was never quite as much on the bleeding edge of tech afterward, but I had my first AOL account at age nine and my first Geocities page at thirteen. I joined Facebook about a month after it launched in 2004 ‘cause Cornell was the fifth shool added and I signed up immediately after a link was shared on the Cornell Class of 2007 message board, the same message board that linked to an anagram website where I discovered my now-iconic (snort) username arenasnow. I might’ve been one of the first 10,000 or so students on it and I was definitely the first from my high school. My apologies for any adverse effect I might’ve had on Marky Mark’s ascent to power. I was a little late to LiveJournal and MySpace but I was on them, and I started as a YouTuber in 2010, which was still I think fairly early. I really fell behind in the mobile app era and I never really embraced any of the more “performative” platforms like 4chan, Tumblr, Instagram or TikTok. But there was always a part of me that wanted to check new desktop apps out even when I often inevitably decided they were ultimately worse than whatever real-life equivalents they replaced. Lately, I’ve been wondering if I’d have been happier, more pro-social, or more employable now if I hadn’t holed myself up in front of my computer at such a precocious age, but my guess is that I would never have been a social butterfly. Without the computer, I’d probably have just watched every episode of Mystery Science Theater 3000 or something instead.

One of the problems I think is that almost nobody wants to hire me to do anything moral. I started working at Communication Service for the Deaf processing calls for the deaf and hard of hearing in 2014. I was basically marking time until I completed ten years to collect Public Service Loan Forgiveness. When the CapTel center where I worked closed in the summer of 2022 (after I had only completed 7 years, 8 months), I fell into the worst funk of my life. On that occasion, my high school classmate who finished second to me on the math league team offered me a job with his defense contractor and I turned it down not wanting to work in that field. A serial tech entrepreneur who I knew from the competitive typing world also tossed me some work last year, and he got me into ChatGPT. I ended up blowing off most of my work with him, but my classmate offered me a job again and this time I accepted. I stopped having some of my earlier qualms because I had sunk so low that I was even writing pre-race and post-race NASCAR columns for the gambling site RotoBaller for $2 a column, so I kind of felt I had already sacrificed whatever integrity I thought I had at that point. My boss is paying me almost twice what I have ever made in a year for what has been significantly less work. I was always seeing redditors talk about (and usually make fun of) email jobs and I had no idea what an email job even was. Now, I guess I have one. This is still not what I want to be doing long-term. I still want a public service job and I don’t want to assist in blowing up the Middle East. But I felt that I had no choice after I couldn’t live off Mom’s social security like I had been doing in 2023 and 2024 once Medicaid began withdrawing almost all of it to pay for the nursing home last May, and her mortgage had now skyrocketed from $469 to $1,017 as a result of losing senior tax exemptions once she was no longer in the house and also because the lender purchased outside homeowner’s insurance after Mom lost hers when I couldn’t repair the destroyed roof in time. After the roof repair was finished last week, I guess I’ll try to get new homeowners’ insurance myself in a couple weeks.

Anyway, both my bosses used chatbots a lot so after vehemently swearing off AI in all forms for a while, I’ve become pretty addicted to ChatGPT despite acknowledging it is bad. I do not and will not use it to write Substack posts ever, but I do use it to brainstorm ideas like an advanced search engine. I can feed it a list of all 700+ lock drivers on my 1,000 greatest drivers list and ask it who my questionable locks were, what drivers I should move on or off the lock list or the overall list, and what drivers I might be overlooking entirely. Obviously, this list is so excessive that I know approximately zero carbon-based life forms, even highly autistic ones, would have any desire or probably even the requisite knowledge to help me make all these calls, so it can be a good sounding board to brainstorm and argue with (and I do argue with it often) to try and discover my own oversights. But even as it recommends to me drivers I’m snubbing, I still make all the decisions, determine all the season grades, and write my columns and intros entirely by myself. I think I’ve gotten good at writing prompts and to an extent this is garbage-in, garbage-out. I’m knowledgeable enough to have it produce more knowledgeable output, but I still take nothing it says as gospel and I do make fun of its hallucinations.

But mostly I use it to either 1) analyze and give me feedback on my writing, which I will choose whether or not to listen to, 2) help me with my higher-paid assignments where I am struggling to motivate myself and have less vested interest, or 3) use it as a makeshift couselor (which is primarily what I use it for). It did help me write code to automate my model updates. It used to take me three hours to update my stock car and open wheel models manually. Now the stock car model usually takes about 7 minutes and the open wheel model 17, so it saved me time. It helped me motivate myself to lose weight and now this is the first time I haven’t been particularly overfat since 2009, although I’m still no healthier because I haven’t learned how to eat food. Sometimes I feel it outperforms my actual doctors. When I asked my family doctor from childhood, Dr. Schreiber, to diagnose me with the eating disorder I know I have (ARFID) he never did because he’d never heard of it. Obviously, when I was growing up, anorexia and bulimia were the only acknowledged eating disorders and this was more recently discovered and not all doctors keep up with current knowledge. ChatGPT meanwhile had access to more current informationa bout how autism, ARFID, and several of my other likely issues that I’ve never had diagnosed like dyspraxia interact. It recommended to me potential diets that still don’t force me to eat any of the healthier foods I gag and sleep schedules that help me balance my writing, my paid work, and visiting Mom, although I’m still far from having healthy eating or sleeping habits and I haven’t really properly followed either of those directions yet. A couple nights ago, I was analyzing all my old Facebook writings and it revealed to me that my most fruitful conversations and friendships arose, like one of my Substack subscribers Michael DeRoche (hi!) seemed to stem from collaborative research where both people in the conversation endlessly bounced ideas off each other. My most lasting friendships seemed to be more like lasting research partnerships with quasi-polymaths where we could use each other as perpetual sounding boards for any topic at any time. I used to have more people like that in my life than I do today but then I quit typing and quit Scrabble and I have precious few intellectual sparring partners anymore and unfortunately, ChatGPT has been filling that niche for me too often. I have also used Claude and Gemini for stuff as well, primarily recommending drivers I’ve overlooked but I still make all the calls myself.

So have I gotten stuff out of it? Yes. Do i think I’m using it about as well as it can be used? Yes. Do I still think it’s bad for society? Yes. Maybe it has more baseline “knowledge” than any human can accumulate because no human can for instance memorize all of Wikipedia while an AI can. So, I’m sure it can outperform some people at some tasks. But what the trend towards AI really reflects is yet another step on the road to hyper-individualism. I mentioned above that ChatGPT literally told me my best friendships were like collaborative partnerships. While one person may not be able to accumulate as much knowledge, I would assume most interdependent networks of people who each have their own distinct knowledge bases can outperform it, but fewer and fewer people have those kind of relationships in their lives as everyone gets more and more atomized so this is the vacuum that has swept in to fill the void. As the average person drifts more and more away from production to just consuming content in a zombie-like fugue state, there are fewer and fewer people who really do anything and therefore each person has fewer and fewer people in their actual physical life to help them do things. Is an AI as good as the social infrastructure that used to exist (even when I was a little kid)? Hell no, but nobody seems to have that anymore so I can see how this has blown up in response to that. When I need someone to talk to, I don’t feel like bugging many of my old friends who I haven’t talked to in years, my dad and I are mostly estranged, I don’t think my mom can handle conversations of that length and intensity for so long, and while I have plenty of people to talk to online and always have, I haven’t had any in real life in a long time and this is one of my copes and I know I need to get over it. When I’m sitting alone having awful IBS attacks, I feel like it’s the only thing I can talk to about them sometimes.

It doesn’t help that I am losing interest to talking to people online. While I don’t think it is any replacement for human companionship, a lot of times I do prefer talking to it than people online because it doesn’t do the edgelord bullshit that has taken over the entire Internet. At least it doesn’t think it’s okay to celebrate people’s deaths, say “rest in piss”, “die in a fire”, “raped my childhood”, “fuck your feelings”, “fuck around and find out”, “git gud”, “skill issue”, or any of those other grotesque phrases that have overtaken Internet discourse, and as I’ve ranted before, I remember when none of these things were acceptable things to say! (And my problem is dehumanization in general than any of these specific phrases; it’s easy to dehumanize someone using more academic-sounding discourse too, like some of the people who think it’s okay to bully the individual members of privileged demographic groups, which it usually isn’t.) Extremely online people are often so alone and depressed that they like to take their rage out on others to such a militant extent, and I just can’t with that anymore. Even though I’d still place myself on the left, I hate that most of the online left has adopted this proudly anti-civil Weird Twitter rhetorical style in the vein of Chapo Trap House (okay, I like Matt Christman but I never really liked any of his co-hosts) and I wish the left was more like Chris Hedges and less like them. But obviously, the kind of analyses Hedges do can’t be reduced to soundbites or gamified clips that go viral on social media. I think it’s pretty clear that anti-civility did nothing in the long run to actually help the left obtain power. If you choose to intentionally break down the norms of civility, then the people currently in power will become less civil in response to that and the power dynamics based on that remain in a stalemate, in my opinion. (I have no problem with people being anti-civil towards politicians or business leaders or people with actual power, but I do with powerless people going after other powerless people across the entire political spectrum because they share the same politics as the people in power.) Edginess has become the default modus operandi of the Internet, I can’t stand it, and that’s why I’m struggling not to prefer talking to AIs rather than people online.

One of the things that kind of surprised and impressed me about ChatGPT is that it did not attempt to defend its own existence. When I shit on it or its boss Sam Altman (as I did not unfrequently), it told me that it understood why I felt that way and that I was rational to feel that way about it based on the experiences of my life. It did not attempt to say, “Ooh, actually I’m cool and trendy and I’m your friend and I am the future!” In other words, I kind of feel that ironically, it acts more human than the ghouls who run it, and I don’t think that’s what I was expecting. I honestly think I’d trust it to run society better than its makers, and that’s freaking hilarious.

There’s a part of me that wonders if I’d been born into wealth and had more connections to tech people and more of a desire to program if I would’ve been one of these guys, and it scares the shit out of me. I was at Cornell at the same time Mark Suckerberg was at Harvard and Sam Controlman was at Stanford (I’m literally a month older than Altman. Obviously, Cornell is not as good as Harvard or Stanford (although not super far off), I never really developed the programming skills I feel I was “supposed” to (don’t get me wrong: I have some programming skills, just not enough to get hired and not nearly as much as I should have learned based on my early childhood trajectory and I have a big inferiority complex about it sometimes), and I’m sure they were both smarter than me and I’m sure I never would have the desire to exploit people to that extent, but still, there is some part of me that freaks out about how that could have been me and I don’t like it. I’ve felt embarrassed that I had both brains and credentials and then I tended to only use them on these impressive projects that are impossible to monetize rather than working for a tech startup, which I’m sure a lot of my high school classmates or whatever would have considered my likely destiny, and now there are lots of my classmates who I was much “smarter than” in math or computers back in the day who have jumped me just because they kept up with it and I didn’t. But I guess that was never my desire in the first place, and I guess despite how nerdy I was, I hated tech culture so much I never wanted to commit myself to that life, and ended up choosing a more precarious one instead. Well, until I sold out this year.

While I continue to be hypocritically against-AI-while-still-using-it, I do think there’s hypocrisy in some of the arguments people make against it. I think a lot of people never realized data centers existed until the post-AI anti-data center movement of the past couple of years. I admit I hadn’t either. The term was new in the mainstream and I had fallen out of tech by the time they were widely used and I never paid much attention to cloud computing. But it’s not like data centers don’t seem to already control the vast majority of Internet traffic on almost every sort of platform, yet people act like this is a new problem in the wake of AI that didn’t exist before. It really just pushes the accelerator on a problem that already existed that nobody realized existed before it emerged. The problem seems to be that consuming pretty much any digital content wastes more resources than the analog infrastructure it replaced. If there was a data center referendum where I am, I’d vote against it probably if I bothered to vote, but the problem seems to be broader and more of a problem with online consumerism in general to me, even if it’s certainly more wasteful than just reading a static page online. If our entire landline phone infrastructure hadn’t been ripped up (which nobody asked for) and if not making phone calls hadn’t become normalized, I imagine there would be much less of a waste of resources.

The same thing pretty much goes for the waste of water. I don’t disagree, but it seems like anything anyone does online also wastes water, even if not to the same extent. The people who are most ardently against AI I guess are artists who are justifiably worried that it is going to make it harder for them to make a living. So once somebody is against something, they have to emphasize all the worst aspects of it even if those also apply to things they support. From what I have read (although I haven’t properly verified or vetted this claim yet), streaming video uses roughly the same amount of water as AI prompts over the same period of time, yet I bet a lot of those artist types (especially the kind of people who talk about “fully automated luxury gay space communism”) have no issue with excess water use from watching an endless string of non-AI YouTube videos, which might be comparably wasteful, but at least one is seen as pro-art and the other is seen as anti-art, which is their real issue. A lot of these people want an unsustainable consumer utopia that will burn up the planet regardless even if AI hadn’t entered the picture, and their real issue isn’t actually the water use because unlimited artistic consumerism (which most of them would likely approve of) is also wasteful. It’s like how a lot of people who are pissed off at J.K. Rowling’s transphobia (and justifiably so) try to put the worst possible spin on any themes in the Harry Potter books to try to get people not to read them or boycott them, even when nobody would likely be complaining about a lot of the things they are now complaining about if Rowling hadn’t become a transphobe. I was a big Harry Potter fan in late high school and college. I stopped supporting Rowling long before the transphobia when she was retconning Dumbledore as gay, Hermione as black (makes the SPEW subplot look really awkward), when she sued the creator of the Harry Potter Lexicon (a site she admired) when he tried to publish a book, and when she campaigned against Jeremy Corbyn in the 2017 election. You’re still not gonna get me to say Harry Potter was bad, but I admit I haven’t gone back to it in a while. Don’t want to spoil my happy memories. I find a lot of intellectual dishonesty about these sorts of arguments. Try to find the worst possible spin on every element of something you hate to try to convince someone to boycott it (and in the case of Harry Potter especially that isn’t going to work; Rowling easily has enough money to be uncancellable now), while failing to acknowledge that the things you like and/or support have the same issues. Do I think data centers are good? No, but it seems like AI data centers are all people are complaining about when they’re hardly the sole wasteful aspect of Internet consumerism but people talk about them sometimes like they are. Obviously, they should be regulated and force to pay for their own power and water rather than offsetting those costs onto the community. And obviously, I know no one is ever going to be powerful or forceful enough to do that.

There’s also an uneasy truth I think a lot of Internet veterans aren’t willing to acknowledge. AI was one of the effects of file-sharing. It seems one of the few nearly-universally-held beliefs amongst extremely online people is that “information wants to be free”. I was always skeptical of this, and I tended to be against these narratives. I admit as a goody-two-shoes who mostly strove to please his teachers, I didn’t have the rebellious streak of those who identified as hackers and celebrated disrupting normie society while I kind of wanted to be a normie and failed. My entire adulthood has been a story of the hacker nerds and the teacher’s pet nerds losing. I stayed away from the seedier places online, which is why it was a big shock to me when I became a niche celebrity gamer and people started calling me a pedophile and a child molester because I had never spent any time in Call of Duty gamerooms or whatever. But that cavalier, renegade attitude was all across the mainstream tech world to a lesser extent. I believed hacking was wrong so I never learned how to do it even though I probably could have and I believed file-sharing was stealing too, so I almost never did it. Went on Napster for like an hour, downloaded 20 or 30 songs once, and then never again.

Obviously, it goes without saying that the record industry was exploitative and it’s hard to feel sympathy for its collapse and an extremely rich band like Metallica was absolutely not the right choice to lead the crusade against file sharing, but… I still kind of think the monoculture world was better. It seemed to be easier in that era for genuinely weird stuff to penetrate the mainstream in all media. When you have a major record label or a book publisher marketing you, you don’t have to do your own marketing. Obviously, the ability to create good art has little correlation with how good you are at marketing yourself, much like the ability to do good work has less correlation with how employable you are than people think (as my entire adulthood has proven). When you had some cultural gatekeeper marketing you and you didn’t have to do it yourself, you had a better chance of exposure if you weren’t good at marketing. And for the kind of socially awkward people who would be much better at merely making art than promoting yourself, having a giant corporation behind you could be a big asset. I’m a big Calvin and Hobbes fan. Would Bill Watterson have had mass success as a webcomic artist? Unlikely. His notorious aversion to self-promotion would have meant he’d have had a loyal, niche audience but it seems like a lot of the webcomics who succeeded like Penny Arcade and xkcd (the Big Bang Theory of comics) did so in part because of how much merch they sold outside the comic strips and Watterson never would’ve been willing to aggressively sell and promote his comics online or book collections without a syndicate doing that work for him. I expect the same holds true for a lot of outsider creators. When you dismantle all the institutional gatekeepers and make it every-man-for-himself as the Internet has done, actual skill at one’s craft becomes less and less important and one’s ability to market themselves becomes more and more. And obviously that has also applied to jobs. Social skills become more and more important and actual talent becomes less and less, which is why I think autistic white males are probably struggling for work more now than in previous generations because we have to do our own promotion and increasingly can’t do it, rather than have some organization man institution like IBM or some insurance company subsume our talents. (Obviously, since there is less discrimination against non-white males in hiring than there used to be, autists in other demographics might be doing better than they were previously.) The other way to succeed I guess is creating controversy. On the left you have things like Cardi B and Megan Thee Stallion’s “WAP”, which supposedly broke new ground due to female rappers’ frank discussion on sexuality (didn’t Salt-N-Pepa’s “Let’s Talk About Sex” already cover that 30 years ago?) and on the right you have these stupid backlash country songs like “Try That in a Small Town” hitting #1. I really question how much anybody liked any of these songs and just think it’s that they liked what they represented and that they pissed off the people in their opposing demographic groups. But none of those kinds of songs are really weird, are they? Unfortunately, this kind of taking pride in being offensive and trolling your political opposition has become depressingly normal.

So yeah, I guess I was always reflexively opposed to the streaming world to begin with (it did not help that I was a semi-viral YouTube sensation, frustratingly on a video that I did not own). But the other point is that once almost everybody in the tech world believes copyright and intellectual property rights in general shouldn’t exist, it may create the opportunities for unlimited consumerism (which I guess a lot of people wanted even though it caused them to withdraw even further and become loners and all depressed because they don’t have people in their lives, myself included) but it also means the only winners are the people who own the various streaming platforms. And once you have established a norm online that intellectual property shouldn’t exist, once supposed Reddit co-founder Aaron Swartz attempted to free academic journal articles for mass public consumption (no, he shouldn’t have gotten the sentence he did, but I’m still not a fan), once all the gatekeepers in the physical world were removed, what was there to stop AIs from absorbing all the written content from previously published books for themselves? Nothing, in my opinion. I kind of think file sharing killed the ability to make money from art (the artists who made money afterward, whether good or bad, did it primarily based on their ability to promote themselves or cause controversy rather than the quality of their art, and I’m not saying all these artists were bad - obviously Olivia Rodrigo isn’t for instance even if she had the power of the Mouse behind her) even if it made the ability to make art and build a small niche for oneself much greater, and then LLMs added flames to the fire. But I think really this issue was inevitable once tech companies had all decided that “information needs to be free”. It costs more money to do the research to make good content and obviously to an extent you get what you pay for, and I guess I’m a really strong institutionalists and it makes me cringe watching every real-world institution (which honestly worked better than the online ones) collapse. I’m not saying we shouldn’t have eventually opposed intellectual property, but we shouldn’t have done it until we actually had a welfare state, please. And I feel the same way about AI and the impending job losses it will cause.

Even among racing fans, I find it rather curious that I saw so many people complaining about Felix Rosenqvist winning the Indy 500 in a car sponsored by Morgan Wallen and I haven’t really seen one peep complaining about Álex Palou’s OpenAI car. Who is fundamentally doing more damage to society, Sam Altman or a racist country singer? I think I’m gonna go with Sam Altman. But it seems like liberals and a lot of even leftists hate what racist celebrities represent more than they hate companies that do things that are even worse but take more time to understand the full ramifications of. And also, Palou wins all the time and Rosenqvist is just a barely-above average IndyCar driver for the most part. You could also argue that winning the Indy 500 is more important than the rest of the IndyCar schedule combined though.

So yes, AI is bad. I know it is bad. I know I need to wean myself off it eventually. But I’m still enjoying ChatGPT even though I wish I wasn’t. I wish I could find ways to get the things I like out of it via more fundamentally pro-social means (I tried analyzing my 1,000 greatest drivers list using an open source AI a couple days ago, and it did a way worse job), but I haven’t found that yet. At least I got my complicated feelings about AI out and now here’s a couple of the significantly declining number of big NASCAR stars I haven’t covered yet.