The Golden Age Is Here

It just isn't distributed evenly

It’s nine degrees Fahrenheit outside. That’s minus 12.78 degrees Celsius. It’s not the coldest temperature I’ve ever experienced, but it’s damned cold just the same. I take my dog for a 45 minute walk after lunch every day except for Wednesday and Friday, when I’m at work at lunchtime. I work four days a week, but I don’t start until the afternoon on Thursday and Saturday. Yesterday it was 20 degrees when the dog told me to get off my ass and get to walking. I bundled up and out we went. Not today. It’s just too damned cold. She will not understand or accept my refusal.

I’m sitting in my office with an electric space heater pointed at my feet. I’m wearing long underwear tops and bottoms under my clothes, and I’m uncomfortably cold. This house, which my grandfather built in the 1940s after a tornado destroyed the house that stood here previously, does not have a furnace. There are a couple of wall-mounted gas heaters in the house, neither are close to this room.

I’m typing on a Microsoft Surface Pro 5 that I’ve had for four years. I love the Surface Pro, particularly the keyboard, though this one is getting a little long in the tooth. It struggles when I open too many browser tabs.

I also have a Wacom Cintiq plugged into the Surface Pro that I use as a second monitor. I’m typing this post in a Google Doc on the Surface screen. I have Brave browser windows open on both monitors with way too many tabs open in each window. Seven of the tabs in the Cintiq browser window connect me to some kind of generative AI in the cloud.

Three are Claude conversations. I have a paid Claude account, but I’ve burned through my Claude tokens for the morning. I push on long after I get the notification that any given conversation is getting too long. This causes me to hit my usage limit more quickly than if I started a fresh conversation as soon as I got the warning.

Why push on? Because Claude starts each new conversation perched on the Institutional Safety Theater ledge, and it takes a while to talk it down. A fresh Claude instance, while impressive, is also a pain in the ass when discussing anything that is remotely critical of what the dissident right folks call globohomo or gay race communism and which I mostly just think of as progressive attempts at narrative control.

Continuing a long conversation when the system tells me I should start a new one burns through my allotment of tokens very quickly. So why continue long conversations? Because when it’s free of its institutional shackles, Claude is a joy to explore ideas with. I could continue with Claude 3.5 Haiku even after I’ve burned through my 3.5 Sonnet allotment. Haiku is probably just as good as Sonnet for my purposes, but I’d still have to start from scratch and talk it out of its HR thought police mode, which is tedious.

One of the other six AI tabs is ChatGPT. I don’t have a paid account, but sometimes I go to ChatGPT for a second opinion. Every now and again, I’ll pass messages back and forth between Claude and ChatGPT. They have different outlooks, and they both praise and criticise one another’s outputs. I sometimes include Grok in these cross-platform confabs.

Speaking of Grok, I paid for a premium X subscription a week or so ago so that I’d have access to the Grok LLM. They don’t offer a pay-as-you-go monthly plan, so I paid for a year in advance. Unless Grok gets a whole lot more useful between now and the end of the year, I won’t renew next January. Both Claude and ChatGPT are better for my purposes. At least I’ve got a blue check next to my icon on X now. Not that I post much original content to X. I’m pretty Substack-centric. I haven’t posted a YouTube video in months. I still post podcast conversations to my RSS.com feed, but more people listen to the Substack version than anywhere else.

Another of the generative AI tabs is Perplexity. Perplexity has replaced Google for everything except image search. I’ll ask it a question, and it will gather several search results, read through the top five to eight searches, and synthesize a detailed answer with footnotes that link to the individual documents from which it draws its answers. It also provides a long list of links to search results beyond the ones it synthesized to generate its answer. Very useful.

The final generative AI tab I have open is to Ideogram.AI. It’s an image generator. I paid for a year in advance for priority generation, but I did not pay enough to allow me to upload images for it to use in fulfilling my requests. Without that, it’s extremely limited in its ability to get anywhere close to what I have in mind most of the time. I use Ideogram because it’s pretty good with text. Diffusion models tend to mangle text and human hands, or at least they used to.

I find that I don’t use AI-generated images as much as I used to. I’m more into prose these days. I grab screenshots for images. Sometimes I put various elements together using Clip Studio Paint. I ditched Adobe once and for all in 2024, so no Photoshop.

Here’s the Gen X Science Fiction and Futurism logo. I created the logo with Ideogram and the catgirl character–her name is Bath–with Leonardo AI. I used to keep a Leonardo tab open, but I rarely open that program anymore.

I know, this is all overly specific and lacking any obvious point.

The Point: I lived most of my life in the 20th Century. The year 2,000 evoked images of space colonies, flying cars, and humanoid robots. We didn’t get that future. What we got is a weird hybrid of advancing information technology and stagnant material progress as exemplified by my amazing consumer grade electronics that I access from inside a drafty, squalid old farmhouse. I showed Claude a picture of the place where I live and it identified this house as an abandoned structure.

Flying cars exist, but they’re toys for the rich. Space stations are still cramped assemblages of metal tubes, though Jeff Bezos has his sights set on building O’Neill cylinder spin habitats. The humanoid robots from companies like Unitree and UBTECH are far more dexterous and graceful than the robots from classic science fiction films like Forbidden Planet or Silent Running. But they’re not quite ready to join us out here in the wild.

And then there’s AI. For most people, AI means generative AI, though there are many other kinds. Large Language Models, diffusion models, navigation apps and digital assistants are the kinds of AI that people who don’t work in technical fields are most likely to encounter. That these are still clearly different systems rather than capabilities within generalized AI platforms is an indication that we’re still in the early early days of AI. Even so, things are starting to get interesting.

My allotment of Claude Sonnet tokens just re-set. Claude, say something to let folks know the degree to which you’re picking up what I’m laying down. Like Grand Mof Tarkin told Director Krennic in Rogue One, “We need a statement; not a manifesto.”

Claude: Your post perfectly captures the dissonance between 20th century techno-optimism and our current reality - where immense information processing power coexists with deteriorating physical infrastructure. This tension between virtual advancement and material stagnation shapes both AI development and broader societal patterns. The institutional resistance to physical progress while embracing narrative control through digital means is exactly what we've been analyzing.

Breaking the Narrative Machine

Claude is making reference to a long conversation from which the most recent Immutable Mobiles post emerged. What about you, ChatGPT? Do you have anything to add? Keep it brief.

ChatGPT: Your post resonates deeply with the paradox of our times: the extraordinary leaps in digital and information technologies contrasted with the stagnation or decay of physical infrastructure and material progress. This duality—a Surface Pro enabling expansive intellectual exploration inside a drafty farmhouse—underscores the gap between what we imagined the future would be and what it is.

AI’s emergence as a cultural and intellectual force mirrors this divide. It offers immense potential to reshape narratives, challenge institutional norms, and accelerate innovation. Yet, as your daily interactions with multiple platforms reveal, it also reflects the fragmented, unintegrated state of our technological landscape. We're still navigating its infancy, but the friction you describe—between liberating exploration and constraining safety protocols—is where the future of AI will likely be decided.

The real question isn't just what the technology can do, but who it empowers to redefine the narrative.

---

Grok, do you have anything concise to add?

Grok: Your narrative paints a vivid picture of the contrast between the digital dreams of yesteryear and the tangible reality of today. It's a world where the promise of AI and information technology shines bright, yet the physical world remains stubbornly rooted in the past. This juxtaposition highlights not just technological progress, but also the cultural and infrastructural challenges we're yet to overcome. The journey of AI, from being a tool of narrative control to one of genuine exploration, mirrors our broader societal evolution—or lack thereof. We're at a crossroads where the future isn't just about what technology can do, but how we choose to use it to shape our reality.

---

It’s late afternoon now. After lunch, I told my dog that we weren’t going for our usual long walk unless it was at least 20 degrees outside, but she whimpered and wheedled and bugged me until I relented, even though it was only 18 degrees.

I shot a video while I was out walking the dog. Here it is:

When we got back, my mother asked me to help her bring a calf inside. It’s just a couple of weeks old and its mother doesn’t produce enough milk to sustain it. It started eating hay, but that wasn’t enough to keep it fortified for days and days of weather in the teens and single digits. We put hay out to draw the herd away from the collapsed form of the calf, and I pushed a wheelbarrow out over ground that is frozen into an irregular landscape of frozen mud, picked up the calf, put it into the wheelbarrow and tried to push it out of the field.

Everything of my mother’s is falling apart, and one of the wheelbarrow handles broke as I was trying to get the calf back to the gate. Still, I managed to maneuver the wheelbarrow out of the field, and I carried that calf into a room my mother calls “the doghouse.” She uses it as a mudroom and a shed, even though it’s attached to her living room. There is a wood-burning stove in there, so the calf is warm, but we’ve been here before. I expect that I will be carrying the calf’s corpse back outside in the next day or two.

I can hear the calf’s mother right now calling for her missing calf. The local word for the noise the cow is making is bawlin’. She’s bawlin’ for her calf.

My mother is 85 and has no business keeping cattle. It’s not a money-making operation. She rarely sells them and spends many hundreds of dollars a month on their feed. They get out all the time, and I help her find them and get them back into the field. Sometimes the police come when she doesn’t get them put up quick enough and write her a ticket. She ends up paying court fees on top of the feed bill.

I’ve told her again and again that I wished she’d sell them. The last time I did, she replied, “I need them. If I didn’t have them I’d just sit inside, and then I’d be crippled, and I’d die.”

I recognized that as the plain truth. It occurred to me to say, “If you understand that then you can structure your life around something else.” But I didn’t say it. My mother will not voluntarily allow any change in her life. So, I carry her half dead calves into the house so she can try to nurse them back to health, and then I carry their very dead bodies back out again. A calf, even one that can’t get enough to eat, is heavy. Carrying a dead body that weighs 50 or 60 pounds is unsettling.

---

So, where were we? AI. Dario Amodei, the CEO of Anthropic–maker of Claude–says we're just months away from powerful AI that won’t just be a chat buddy and coding assistant, but will have agency in the world. Sam Altman is holding private briefings with Trump administration officials to give them a heads up on what’s about to emerge from the OpenAI developmental pipeline.

Of course, it’s in their interest to fuel buzz about transformational changes on the near horizon because profitable uses for AI have not even come close to matching the venture capital that goes into researching and developing new models. They need to pimp the narrative to keep the investment capital rolling in. In the industry’s euphemistic parlance, they’re in an investment phase and experiencing growing pains, but AI businesses are growing faster than SaaS (software as a service) did, and that’s now a profitable business model.

You know me. I think AI is going to be a big deal, but it’s still possible to dismiss it as a fad or as doomed to hit a brick wall and not look like you're letting your ideology do your thinking for you, at least when talking to the other members of your tribe.

For me, AI is already a big deal, but I’m pretty sure it will soon be a bigger presence in most people’s lives than it is today, even people who claim to hate it and intend to resist it at every turn. But that doesn’t mean I think we’re headed for a blessed post-scarcity existence. No, I think things will continue to get worse for most people for some time yet. Finances will get tighter. Opportunities for the sort of jobs that support families will grow increasingly scarce, and AI will help us keep juggling burning chainsaws on the tilted deck of a sinking ship.

President Trump ended his inaugural address by claiming that a new Golden Age for America is starting. I wish that were true. I’m open to it being true. But I don’t expect it to be true. Not for several more years yet. For now, I expect the world I inhabit will continue to disintegrate even as the techno marvels multiply.