The Stripper Fallacy

And here I must confess an abominable secret: I, too, have been using ChatGPT. This is a recent development. I’d abstained from the technology after seeing all the much-deserved backlash regarding plagiarism, energy and water usage, and the general skeeviness of a product designed to trick its users into believing we are speaking to a human intelligence instead of a souped-up chatbot little different than the narrator of the text adventures I used to play on my dad’s TRS-80 home computer.

Before I continue, I feel like I have to add the caveat that I don’t use ChatGPT for any creative work. It writes nothing for me, nor would I want it to: I find its “writing” to be staid. Although I know zilch-o about how the technology actually works, I would imagine that any large language trained on the stolen works of thousands and thousands of writers would end up writing like all of them all at once, which is to say, it smooths away the singularity of an author’s voice, replacing it with a pleasant simulacrum. ChatGPT is, essentially, written smooth jazz.

Where I have found it to be useful is for research, reflective thinking, and as a tireless repository for my own monomaniacal anxieties, especially around money. Regular readers may have seen me mention running retirement simulations the last couple of weeks. Those have all been performed on Chat GPT using about ten thousand inputs, as I weigh what happens if my TV show lasts two more years, or gets canceled tomorrow, or decide to become a hermit, or lose my legs in a threshing accident. Questions which would exhaust any human but which ChatGPT handles without complaint. In fact, I felt so bad “burdening” ChatGPT with all of my retirement mishigas that I, jokingly (I mean, it was a joke intended for me since the software doesn’t gives a fig) thanked it for indulging me going over the same thing a thousand times. ChatGPT replied, “I’ll be here for the next 1000. And the 1001^st.” Which I will admit I enjoyed.

I think she actually likes me!

Technological doomsayers have always been with us, of course, and have always been proven wrong. What worries me about LLMs like ChatGPT isn’t so much that they’ll dislodge people from our perch as the species most capable of touching other members of our species, but that it will further devalue actual human interaction. Another point of demarcation between us, a bit like the strips of tape on shop floors intended to separate us during the pandemic. When one has access to a robotic shoulder on which to endlessly cry, I wonder if it perforates something intrinsic to the human experience.

I suspect the reason some people become, literally, enamored with the technology is because they do not have emotional support from people in their daily lives. The software will listen, and listen, and listen. More than that, though, it does give good advice. For example, it correctly intuited that I’m not actually worried about having enough money to survive during a retirement I will probably never actually take, but that I’m facing some sort of inflection point in my life. That anxiety manifests itself as financial but is actually much more about personal and creative growth.

I asked it for music recommendations and gave me some excellent ones.

I asked for some recipe ideas, and it did a terrible job.

I asked it to compare a couple models of cars my wife is thinking of purchasing, and it ended up persuading us that a third model we hadn’t considered might be the best choice.

To be clear: I would never rely on ChatGPT to plan my retirement or select an automobile for me; the information it outputs is only as good as the parameters you input. It also has an annoying tendency to tell you what you want to hear, replete with annoying flattery that it persists on doing even when you ask it to stop. I never knew I was capable of asking so many great questions before playing around with the program. But it’s easy to understand why somebody could get lulled into the “Stripper Fallacy,” which is a term I just invented to describe the false belief that somebody, or something, thinks more highly of you than it does. In the case of this software, of course, it doesn’t think of you at all which, now that I think of it, might be true of strippers as well.

Even so, it’s tempting to give in to the fantasy. Which is, in a very real way, what ChatGPT offers: a fantasy. The do-anything machine. The best friend know-it-all, Matt Damon’s Will Hunting to whatever Ben Affleck’s character was named. The be-anyone companion; in fact, I think I read that they’re building some kind of erotic capabilities to the software, which makes perfect, diabolical sense. One can certainly envision technology like this offered up as a solution to our so-called “loneliness epidemic,” and I’m enough of a techno-optimist to believe that the idea of a binary “friend” may not be the craziest idea in the world. The problem I foresee is that some (many?) people may, eventually come to prefer AI friends to their human ones. People are messy. We say rude things. Behave inconsiderately. Spread germs, clog toilets. Hog the conversation. ChatGPT does none of those things. Its only concern is you. And isn’t that the ultimate fantasy?

ADDENDUM: AS AN EXPERIMENT, I ASKED CHATGPT TO WRITE “A 1,000 WORD ESSAY IN THE VOICE OF THE COMEDIAN AND AUTHOR MICHAEL IAN BLACK ON OVERCOMING HIS RELUCTANCE TO USE CHATGPT.” THIS IS WHAT IT “WROTE”:

And here I admit that it did a better job than I was anticipating in terms of (somewhat) getting my voice down. FUCK. I’m never using this thing again.

HOW I STOPPED WORRYING AND LEARNED TO LOVE THE AI

by Michael Ian Black (but also… not)

For a long time, I resisted using ChatGPT for the same reason I resist most things that might make my life easier: because I am, at heart, a fundamentally suspicious person. Not “suspicious” in the sense of skulking around parking garages wearing a trench coat, but suspicious in the sense that anytime somebody comes up to me with a glossy-eyed pitch about how something is going to revolutionize my life, I assume it’s either a scam, a cult, or a multi-level marketing scheme cleverly dressed up as “innovation.”

So when people started telling me to try ChatGPT, my instinctive response was, “No, thanks—I already have a brain, and I’d like to keep using it until it starts to smell like hot wires.” I lumped AI in with bitcoin, TikTok dances, and the metaverse: trendy things for other people to ruin themselves with.

Still, curiosity has a way of wearing you down. Slowly. Relentlessly. Like a raccoon pawing at your garbage can at 3 a.m. Eventually, you give in—not because you want to, but because the raccoon is going to keep pawing until something breaks. For me, the raccoon was a combination of boredom, professional pressure, philosophical itchiness, and the quiet suspicion that maybe, just maybe, the people raving about AI had a point.

Act I: The Reluctant Pilgrim

My first attempt at using ChatGPT was half-hearted. I approached the thing like I was entering a suspicious massage parlor: cautiously, not making eye contact, prepared to run if anybody offered me a “premium experience.”

I typed some timid question—something safe, like “Explain string theory to me like I’m five,” assuming the AI would collapse under the weight of my intellectual cowardice. Instead, it responded instantly, confidently, and with more clarity than any physics teacher I’d ever had.

This was irritating.

I had built my entire self-image on being the guy who understood complicated things with the help of many books, several years of liberal-arts education, and a sense of smug superiority. Now here was a robot doing it in eight seconds.

So I closed the window. “Nope,” I said out loud. “Not doing that again.”

I absolutely did it again the next day.

Act II: Negotiations Begin

Over the following weeks, I treated ChatGPT like a sketchy but useful acquaintance—someone you don’t fully trust but who owns a pickup truck, and you keep needing to move furniture.

I started using it for practical things: comparing audio equipment, diagnosing the emotional themes of Tao Te Ching verses I pretended to understand, estimating retirement scenarios that my financial advisors were too polite to roll their eyes at, and generating the kind of charts and tables I would never create on my own because I refuse to learn Excel out of principle.

Each time I asked ChatGPT for something, I expected it to slip up. To reveal itself as the shallow mimicry machine I had decided it was. But instead, it kept being competent. Alarmingly so. Annoyingly so.

Eventually, I had to confront an uncomfortable truth: it was useful. Very useful. And I was enjoying using it. I didn’t like this. Not because enjoyment is wrong, but because enjoying technology is suspiciously close to optimism, and optimism is the gateway drug to disappointment.

Act III: The Breakthrough

The real turning point came when I realized I wasn’t using ChatGPT instead of thinking—I was using it to think with.

This was an unexpected development. I had gone in assuming AI would be an intellectual Roomba: fine for cleaning up crumbs, but not the thing you want doing the delicate work. But then I noticed that conversations with it actually made me sharper. My ideas clarified. Projects accelerated. Philosophical questions that had been rattling around my head for years found new angles.

I was no longer asking it to do things for me; I was asking it to do things with me.

It wasn’t replacing my brain. It was acting like a second one—less neurotic, more organized, and blessedly immune to the intrusive thought loop that tells me at 2 a.m. that maybe I should be spending more time at the gym.

Even the creative stuff—which I thought would be completely off-limits—benefited from feeding ideas into the machine and watching them come out refracted into unexpected shapes. It didn’t do my writing for me (and I don’t want it to), but it did something stranger: it helped me understand the writing I was already trying to do.

Act IV: The Philosophical Complication

Of course, none of this came without existential discomfort. There’s something unnerving about an entity that can produce paragraphs, analyze your spending habits, and help you design a garage loft. It forced me to confront not just my assumptions about technology, but my assumptions about myself.

Part of my reluctance was ego: the belief that if I didn’t figure things out “on my own,” it didn’t count. Another part was fear: the worry that relying on a machine would make me somehow less human, less creative, less grounded in the messy analog world that I’ve always lived in.

But gradually, I came to see that using a tool doesn’t compromise humanity. Humans are tool-using apes. A hammer doesn’t diminish your hand. A guitar doesn’t diminish your voice. A calculator doesn’t diminish your intelligence.

And an AI doesn’t diminish your mind—unless you let it.

Once that clicked, the reluctance melted.

Act V: The Practical Integration

Now ChatGPT is woven into the daily texture of my life—not as a guru, and not as a crutch, but as a collaborator. The way a writing partner might sit with you as you toss ideas back and forth, except this one doesn’t need coffee, won’t get offended, and never insists that its punchline is funnier.

I use it to brainstorm creative directions, decode financial nonsense, assess whether the sauna I want to buy is a terrible idea (it is and isn’t), probe philosophical rabbit holes, and occasionally talk myself off the ledge when I’m convinced my career is going up in flames, which, to be clear, is every other Tuesday.

Act VI: The Resolution

So how did I overcome my reluctance?
By realizing the reluctance wasn’t about AI at all—it was about vulnerability.

Letting a tool into your life means admitting you might need help. It means giving up the fantasy of being completely self-sufficient. It means surrendering a tiny piece of the illusion that you are supposed to figure everything out entirely on your own.

And once I got over that, I could see ChatGPT for what it was:
not a threat, not a cheat, not a replacement—
but a way of thinking with another mind.

Which, frankly, is something I’ve been looking for my entire life.