I’ve not written previously about the similarities I see between belief in AI as inevitable and religious faith in God. I have mentioned the idea frequently to friends and colleagues, and several posts here imply the idea. I’ve been meaning to get around to it, but Alberto Romero has beaten me to it with What Apple Knows About AI That Silicon Valley Won't Admit (via Daring Fireball).
AI is like religion. Either you believe it changes everything, or you don’t believe at all. There is no moderate position; nobody believes in AGI “more or less,” just like nobody is “casually religious.” If God exists, the only coherent response is to reorganize your entire life around that fact, as priests do. If you pray sometimes, then you are just an atheist who’s also fearful. When tech companies spend hundreds of billions on capital expenditures to add sparkly AI features to Office, Gmail, and Instagram, I only see fearful atheists—guys who don’t believe in AI but pretend just in case.
It’s this belief in a soon coming, all-powerful digital god that I see underpinning every decision companies and some individuals are making in terms of the actions they’re taking related to AI. I would have said I see more true believers than casual, fearful parishioners, but Romero sees the opposite. He makes a compelling argument.
What they’re doing has a more charitable name in theology: Pascal’s Wager. They can’t prove AI will be transformative. But if it is and they don’t invest, they’re dead. Or worse: the laughing stock. However, if it isn’t transformative and they invest, they only lose some cash they will recoup anyway by doubling down on ads. So they tithe. They go to church on Sunday. Then fast for Ramadan, keep Shabbat, leave offerings at the shrine. They even perform hecatombs, and, every few months, bathe in the Ganges.
I really hadn’t considered this, but I think I might now agree. Please read all of Romero’s post. It’s a well reasoned argument. I’m not sure I agree on all points. I still think Google and its leadership are true believers. I’m also not sure I agree Apple is completely pure in its convictions. But as a generally accurate argument, I think I’m now convinced. I definitely see the fear Romero describes in my friends and colleagues. I remain convinced that one day the tech industry will come to its collective senses and start behaving rationally again. Until then, I remain committed to reading and writing here in the hope that good arguments eventually win out.
I’m writing this while sitting in the Atlanta airport. We’re headed to London for a family vacation. My younger daughter has been studying abroad in Italy, and we’re all meeting up in London. It’s only 3/4th of a family vacation since my older daughter is on a work trip and couldn’t go. Don’t feel bad for her, though. Her work trip is at a fantastic vacation locale. We’re sort of all vacationing, just in different ways.
All that to say, I’m planning on being mostly offline for the next several days. I’ll post if the mood strikes me or if I do some writing during downtime at a pub.
I’m currently reading Ta-Nehisi Coates’s Between the World and Me. I was looking for a short but deep read after recently finishing Transcription. I wanted a book I could finish in a few days, before I leave for a long trip to London.
I’ve had the book on Kindle for a while but recently spotted the paperback in a local bookstore. I’m glad I picked it up. I no longer like using a Kindle—I’m trying to focus more when I read, one book at a time—and also, it’s such a beautiful paperback to hold in hand.
I’m about halfway into the book now. I knew enough about Coates's work and this book to know what to expect in terms of its examination of American history and what it’s like, as the jacket blurb phrases it, “to inhabit a black body and find a way to live within it.” I didn’t, however, expect the book to also be the story of a man becoming a writer.
The writer, and that was what I was becoming, must be wary of every Dream and every nation, even his own nation. Perhaps his own nation more than any other, precisely because it was his own.
This passage snuck up on me. I’ve stalled out here, savoring these words and thinking through their implications. I hope I finish the book now before I leave tomorrow.
I’m back in an Electric Literature phase. In a bad way actually. I discovered this site not too long ago, when I started back to grad school. The sites available now for creative writing are so much better than when I was an undergrad English major.
We’ve been driving through Pennsylvania for almost two months. People give us stunned looks when we tell them: “How long?” When they ask what brings us through, we tell them we’re traveling children’s entertainers, or we’re land surveyors scoping out sites for a new mall, or we’re searching for our troubled niece Paula, she’s been missing since December, have you seen her? Sometimes they offer to buy us vodka sodas, flagging down the kid slinging drinks at the Best Western Plus.
Short stories come in all shapes and sizes. This one has such a distinctive voice. It’s what I really liked about the story. Also, what’s in the trunk? I think I get it, but I’m not entirely sure. That feeling, persisting all the way to the ending, is what makes this story memorable. It’s a story that is somehow of the present day and literary. It’s one time where not being able to put down my phone was a good thing. I was lost in this story. Highly recommend!
I found this book via a link to it on kottke.org where Jason Kottke also noted its beautiful cover. The way the book was described—literary and “a brilliant meditation on those technologies that enrich or impoverish our connection to one another”—I thought, “How have I never heard of Ben Lerner?” After reading it now, I understand how. The book is as much poetry as story, maybe more so even. It’s the kind of book—the kind of writer in Lerner—that I don’t normally like, but this book has stayed with me many moments after I finished it.
It’s still sitting with me, here as I try to make sense of it, read pages over and over again, trying to under what I just read. It’s a thing to behold and wrestle with comprehending. And that’s the best compliment I can pay it.
I have this habit—let’s call it a bad habit—of regularly changing my social media bio. And I mean regularly. I think I updated it like 3 times just yesterday. Partly, it’s that I don’t know how best to represent myself. I like to think I’m not so easily described. That’s a nice thought. The reality is that my interests and expertise are in constant war with themselves. I’m one part software engineer, one part creative artist, and the two are never at peace with each other. My social media profile is the battle field on which that war is waged.
Sometimes I’m a “software developer working on web and cloud.” At other times, I take a more writerly approach. There was a brief period where the bio was “ I am large, I contain multitudes.” See, I told you I like that idea of myself. Thankfully, I realized the pretentiousness of that and deleted it.
Things would be simpler if I wasn’t like this. If I were purely a software engineer and put this writer’s brain to bed, my career would thank me, probably some colleagues too. It’s hard working for large corporations carrying around Hemingway’s shit-detector. If being a writer was enough, that would make for a simpler description. I’m so envious of those 3 sentence writer bios at the back of a book. My engineering side is way too practical to allow it, though. Immediately, my mind starts rationing my bank balance.
After about the 3rd or 4th change yesterday, I settled on this simple form: “software developer and writer.” It’s direct. It’s true. It’s also the form of myself I’m most at peace with. I really do like being both of these things, fully and completely, and conflicted enough to make life interesting. Now, let’s see how long this version of my social profile lasts.
It’s graduation season again, and everyone has advice for you. Follow your dreams. Don’t follow your dreams. Be practical. No, be irrational. Do something people value. No wait, go create something that people never knew they wanted. The cliches are bad enough, but now you’re left wondering, which of these should I follow?
Ignore all this pithy garbage. No one knows what will work for you. Go live your life as best you can and see what happens. You’ll be fine. Or maybe you won’t. No one really knows. Not you, not me, and certainly not that famous movie director who got invited to speak at your graduation ceremony. The mystery, the uncertainty, is what makes us human. Don’t rob yourself of the unknown. Dwell in it.
I’m starting to think wanting to change the world is arrogance. I mean change the world in the capital “W” sense of the word. It looks nice on the surface, but man, the hubris. I would have once counted myself among the group trying, but after reading Jim Prosser’s AI, "Humanity", and Dr. Manhattan Syndrome (Via Daring Fireball), I’ve come to believe it requires a real god complex to actually believe you can do it.
Prosser compares AI frontier lab tech leaders to Dr. Manhattan from Watchmen.
For those who haven’t read writer Alan Moore and illustrator Dave Gibbons’ “Watchmen”, here’s the quick version: Jon Osterman is a nuclear physicist who gets disintegrated in a lab accident and reconstitutes himself as a being of godlike power. He can see across time. He can manipulate matter at the atomic level. He is, for all practical purposes, omniscient and omnipotent. And over the arc of the story, he gradually loses the ability to give a shit about people.
Prosser does a lot to excuse the behavior, trying to take the sting off his message. “It’s not their fault, they just see the world differently,” he would say. Well, not exactly. But it’s close.
This isn’t a flaw in his character. Moore wrote it as the inevitable consequence of operating at that altitude. Manhattan can perceive the entire arc of human civilization. He understands the quantum mechanics underlying all of existence. He genuinely does care about humanity’s survival in some detached cosmic sense. But he can’t maintain a relationship with the woman he loves or comfort someone who’s grieving. Individual suffering becomes statistically insignificant when you’re tracking the movements of atoms and the trajectory of species.
With Dr. Manhattan, I get it. He has god like powers. Tech leaders don’t. I’ll grant that they believe they do, but what would you call someone who believes they’re a god? I think I’m being generous to call it arrogance.
I’ve had several times in my career where I thought I was on the verge of living in the future. It’s a seductive feeling, especially for someone like me who grew up playing with early personal computers and reading science fiction. Eventually, the thing that inspires that feeling fades from public interest or life normalizes around it, and it’s not long before something else comes along to capture everyone’s attention.
I can rattle off a list of these things. The Web, which worked out well for me. Then, Second Life in 2007. That one was done by 2009. The iPhone, another win. Then, Cryptocurrency. The iPad as computer. Even Apple ran that What’s a computer? ad. Somewhere in there I was really believing in smart home automation and voice assistants, which I almost never think about anymore. Finally, there was virtual reality. VR was a big one for me, all the way up through the launch of the Vision Pro.
I sold my Vision Pro earlier this year. I just never used it that much.
Tech people get excited about things. That’s ok. It’s fun for us. Let us have our fun. Just don’t trust us when we tell you something is going to change the world. We’re wrong way more than we’re right.
I was reading Martin Fowler’s Fragments: May 14 this morning. I found this paragraph near the end both inspiring and frustrating.
I’m both a hoper and a doomer when it comes to our AI future. Fundamentally I see any powerful technology as a big bus: we are either on it, or get run over by it. I’m onboard the bus because I don’t think putting up some barriers would stop me being crushed by its wheels. Maybe if I’m on the bus I can join some people to influence the driver a bit. I’m also very reluctant to speculate on the future outcomes of anything, let alone something as powerful as this. Did the early industrialists in the late eighteenth century have any clue what the industrial revolution they unleashed would do? While it created many harms, it also created a massive rise in the living standards of millions of people, at least those whose countries were on the bus. AI may create benefits that I can’t really dream of, although I can glimpse it when it helps a friend stave off Parkinson’s disease.
There’s something poetic and beautiful about this. It’s also a point of view I generally agree with. There’s no way to know the future for anything, let alone a sufficiently complex piece of new technology. The good very well could outweigh the bad. We just don’t know yet, and I think it’s better to be influencing that direction. The trick is finding how to do that without being dismissive of the harms or other people’s concerns.
It’s this idea of being dismissive which is why this paragraph frustrates me. That bus analogy! Ugh. It’s basically the same idea as saying, “Get in on this, or get left behind!” Which is such an incredibly dismissive thing to say. It’s also not true.
No technology is inevitable. Technological outcomes are the results of many factors. Some of it is the usefulness of a technology, but it’s also many people making many different decisions. All of that adds up to an outcome. If someone’s decision is to sit out a given technology cycle, that’s a valid decision. If enough people do that, it will have an effect on that technology’s adoption. We’re already seeing this with the cultural backlash to AI. That backlash is already affecting the frontier AI labs’ strategy and approach.
I’m choosing to be part of cautiously and carefully using this technology. I want to find the good uses and also be a voice of reason against the hype and the fictional stories we tell ourselves about this technology. This bus analogy is one such story that needs push back. This tech isn’t inevitable. It might well be a central part of our lives like the internet—it could also be more like cloud computing than the iPhone—but either way, it will become what it does based on the choices we make together.
