Artisanal Concatenations of Sentient Emanations

Back to school is a special time of year. Traffic gets bad. Everyone gets sick. The usual, you know? Now that we have put that particular piece of calendar indigestion behind us, let’s get back to telling stories.

I have thinned out my list of subjects to write about, and a few of them are the kinds of articles that I have let breathe for a while.

While those are maturing in the things-to-write barrels, let’s bang out an easy one.

Let’s talk about presentations.

When you go beyond basic software engineering as an individual, the communication of ideas becomes important. It doesn’t matter if you are becoming an architect, or a manager, or side-grading into product management; eventually, stuff has to come out of your mouth that makes sense when it slithers into someone else’s ears. If you can match it up to stuff that also hits their eyeballs, bonus points are awarded.

I was afflicted with this early on in my career with a little side quest into enterprise sales. I had to take a 1999-era laptop and build presentations and customer website mockups, often on the plane en route to visiting companies.

Sometime after that, I was blessed with a few speaker opportunities, both domestic and international. Nothing quite prepares you for the moment you are in a standing-room-only room with 2000+ people, and two simultaneous translators into another language, giving a technical talk.

So you should take some time now, before one of these things gets dropped into your hands, to prepare.

Pick a topic at work, or at home, really, and put it into a PowerPoint presentation and practice delivering it to your webcam.

You might think this is silly, but this is a very valuable skill. You will learn how well you can entertain, inform, or persuade a group of people. These are valuable skills to have in a world where sitting and writing code all day long is slowly disappearing. Having built a few decks and business plans with LLM tools, I also assure you that the human origination value is very high in constructing a deck or document.

So how do you get started?

If you are making a basic presentation, I recommend the outline-to-title approach. It takes advantage of one of the oldest communication adages out there.

Tell them what you are going to tell them.

Tell them.

Then tell them what you told them.

Apparently, this bit of wisdom goes back to Aristotle himself.

I leverage this to make my introduction slide, and then I put each line of the introduction slide into its own slide.

As I come to each slide, I will put a few lines of text that prompt what I am going to say; or else, if I am going to be doing something in front of a very fancy audience, I will put a single picture.

I would recommend starting with 3 to 12 lines of text for your first presentations. A 20-page slide deck that has 20 titled images is a masterclass presentation, and it will be a few years before you have refined your skills enough to deliver that.

I will confess that I have stuffed some gimmicks into some of my presentations. If you are looking for an easy crowd pleaser, give away some money. I did a presentation on smartphone games once and I used a series of questions to find the person with the oldest mobile phone in the room. I then took out 100 bucks and asked this person, on behalf of everyone in the mobile games industry and all of the attendees of the show/presentation, to go and upgrade their phone.

I did the same thing a year later and gave a reward to the person with the most mobile games installed on their device for being a shining beacon for us all.

I was poor in year three of presenting, so audiences were simply disappointed that the free money train ended. It was good while it lasted, though.

If you are looking for an easy example of how to throw together a fast presentation, here is an example presentation I gave to a CTO group on how to do creative fractional CTO discounting.  

Putting together a presentation like this at 20+ years into your career can be an exercise in about 1 minute per slide if you have gotten good at it. Because it was for an internet group, there are some liberties with the quality level that work for an internal organization but probably will not hold up when you are faced with a Fortune 100 customer.

Happy presenting!

Hello again, everyone!

It has been really interesting to see how LLM toolchains have evolved from simple image creators to suites of agents swarming over a set of Markdown files that outline a goal. We have seen vibe coding, tokenmaxxing, open-claw, and MCP enter the lingua franca for systems integrators, builders, and customers. The amount of work that people can delegate to the machine layer is staggering, and if you told people in 2020 this would be possible, they would look at you as if you were some kind of insane crazy person.

Speaking of insane crazy people, there is also a lot of literature and science and common sense (are we calling it tradthink yet?) about how much you should be working and what people are experiencing.

“It is rotting your brain,” people say.

“You will never be able to write code again,” people say.

“The people at the frontier are burning themselves out,” people say.

I love coming online and examining each of these statements on their merits. I love to turn them over, this way and that, catching the odd glinting rays off of the sun and seeing their flaws and inconsistencies. I love to look up the speaker and understand the ulterior motive, because there almost always is one, for why they said what they said.

When I am finished enriching my understanding of their assembly of letters, numbers, and punctuation, I declare my own counter-edicts based on my own understanding, experience, and ulterior motives. We will fight fire with fire.

Let’s take LLM-supplemented software development. The average builder today using these tools is producing far more code than they did a year ago, but they are not doing it directly. They are directing tools to issue software, and they may be supplementing that with code reviews and the odd little piece of software here and there that is built by hand. This is the new normal.

If you were to compare that to a human role from 2020, the average software engineer is now doing essentially what their manager did just five short years ago. Understand the shape of the software, the rough high-level approach, and look for the intermittent “aha” that comes from years of experience scaling systems.

There is a subset of people who may have had this role before, or may be ready for this kind of responsibility and take it in stride. The average software developer who has gone through this transition in the past year or two is likely not ready for this kind of switch, and indeed, a good chunk of software developers are not even interested in it. This is one of the reasons why it is creating some burnout. Software developers as a career category have been upleveled wholesale to a software management job with LLM tools. It is a new skill set, and also being applied at an incredibly high rate across the whole software development ecosystem.

The expectation on converting to an LLM-supplemented engineer is that your output at least doubles, and in some cases goes up nearly an order of magnitude. That pressure also contributes to the burnout problem because people want to feel like they are being effective. This is where all of that “tokenmaxxing” nonsense comes in. You have to be driving these tools to do stuff around the clock, and while it is nice to believe these tools will work 24/7 without invoking some sort of union representative to require some level of breaks, they tend to mysteriously shut down or break at odd times. If this happens an hour after you went to scotch o’clock, you are in trouble. You have lost hours of potential work. This means you have some kind of obligation to check in after you have queued up some work to make sure it hasn’t gotten itself into a halt state.

There is one more piece of the puzzle to consider.

The cost of developing software with LLM tools has essentially gone to zero. Adding features can be nearly instantaneous and cost nearly zero dollars. This is a siren song because that means that people might just go and build software because it is cheap or free. That does not imply it is building essential features or valuable items to the platform. This is one of the most dangerous problems people have today. Back in the “old way” of software development, everyone had lots of time for ideas to sink in. They had plenty of days to sleep on feature requests and hold grooming sessions with the creators to make sure the value is captured in the tools and features being built.

In today’s world, it is far simpler to build everything. And that might include building stuff that has zero value, and perhaps items that have negative value. Very few people are strategically looking at the output and saying “we do not need this now” because it is cheap or free.

The problem is that they might make that statement at 11:15 AM. They will be adding features that have marginal value for the next five to nine hours afterwards. If you are as good at basic arithmetic as I am, you will note that this might put you in the twelve hours of work category.

In a world where people feel compelled to build around the clock, check in on late-night agent swarms, and are working at a higher career level than they did just months ago, often with zero new role calibration, is it any wonder that people are getting burned out?

So how can you fix this?

My own strategy is to book some thinking time every week. Make sure you are not prompting around the clock. When you have something that you need to build, think about it. Talk about it. Sing songs about it. Imagine what it looks like naked. Do anything but start building. Every moment you put into the thinking and imaging process, you are going to add value where it belongs, in the ideation layer.

If you feel like you are guilty about this, another thing to do is to do two concurrent projects at the same time that are reasonably distinct. I am always subconsciously grinding away on systems, user interfaces, and scaling problems for anything I am building. If I am building product A, and I also have plans for product B, I can think about product B while building product A or vice versa. Oddly, you might do similar amounts of work on just product A as you would with both product A and product B. You will feel less tired because it is not an overwhelming level of deep concentration on one thing, and you are bringing a fresh perspective to the table very often for each of the projects you are working on.

So if you are experiencing AI burnout, or have a friend who is experiencing AI burnout, this might help you find some balance or sanity in your life. Ensure you are not burning yourself out by going too deep into one thing and having time to think about the right thing to build. Make sure you are “zooming out” enough on the scope of your work, your career, and the habits you are practicing. Think about the gap between what your job used to be and what it is now, and spend time understanding that it is different.

Most importantly? Breathe. If you push too hard, you will get to diminishing returns. You should conduct your time like a symphony orchestra, and have time for the drums, the trumpets, the violins, as well as other instruments. By keeping up with a variety of different things, you will have fresh perspectives and have more time for “Aha” moments and bits of insight that will help spur innovation and new implementation ideas than if you just keep grinding away Sisyphus-style against your codebase.

Thank you for your time. If you wish to support my writing habit, and love to steal nickels from Jeff Bezos, consider supporting this habit with Amazon Affiliate Sponsored Links™! Today, John recommends you look at these two options for great wrist support over the course of your career. I use both a keyboard wrist wrest, as well as one for my mouse hand. I buy these 4-6 at a time because I have multiple computers and also have them handy for travel.

Talk to you all again soon!

Every week, there is a new announcement about which company will win the AI race. This week, all the rage-baiters are posting “Claude is dead” posts and “Anthropic will lose all of its money; KIMI is crushing the metrics!”

Trying to keep on top of who is winning is spinning me around so fast I am getting dizzy.

I want to write a sentence like, “but it is not the model companies who will win, it will be companies who… “ and then I stopped because that is one of those LLM patterns that now triggers me so hard I can’t even complete the sentence.

Now that the spinning and self-loathing has subsided, I want to register some predictions.

This is likely the first “next big thing” race that is going to be won by incumbents.

Unfortunately for the investors in the big models and God knows whatever, this was always going to go to Google or Apple.

Google has been going quietly into deeper suites of tools at the software level, which is where they have a lot of experience and one or two successes outside of search. You might go “ohhh sick buuuurn” and reach for the ointment, but they are not very good at building usable things. I thought the name Google Plus was appropriate for their social software, because everything you wanted to do with it took Plus One clicks compared to a competitor product. That is still true today. They have a giant search playground, a lot of capital, and a lot of patience. They are also innovating in all of the right places instead of greedily trying to suck up their partners’ businesses 100m verticals at a time.

Apple, on the other hand, is making a completely different bet. They are going for the junk. They want to grab the whole LLM market by the short and curlies and squeeeeeze. Their goal is to put a Mac Mini or Mac Studio in every cubicle, bring on-prem LLM hotness to the market, and make the “90% as good, but free” models accessible to businesses for all their fidgety bits and inferences.

These are both good strategies, and oddly, they will work together. Much like the duopoly for the touchscreen device in your pants, there is a very good chance these two companies, one going high, one going low, are going to do this whackadoo pincer move on the startups who are sitting in the middle singing the “I can spell LLM” song over and over, and suddenly the lights are going to get turned out.

Usually the arrival of a “next big thing” heralds a new bajillion-dollar company. I think this is going to be one of the first times that a “next big thing” is going to go to the incumbents.

Does this work well with what is paying your bills today?

If not, then you have been warned.

See you next week!

This is the first blog post I pulled out of a LinkedIn post, I think, but I think it works.

I’ve received more hostility than I expected for being excited to experiment with LLM tools. I’m still working through some of that. Bear with me while I try to articulate why.

To me, LLMs feel much closer to procedural generation than some entirely new category of creativity. They’re another tool for exploring an enormous possibility space—more akin to the technology behind games like No Man’s Sky than the mystical “Gen AI” label they’ve been given.

What I’ve struggled with most isn’t the criticism itself… It’s where the line is being drawn.

If the standard is that we should reject technologies because the companies behind them have acted unethically, harmed creators, or negatively impacted society or the environment, then why stop with LLM companies?

The history of comic book publishers, book publishers, media companies, and IP holders is filled with stories of creators who were underpaid, exploited, or denied ownership of their work. The Industrial Revolution was built on harsh labor practices. The railroads of the American West, the robber barons of the nineteenth century, and the great trading companies of the seventeenth century all left complicated—and often deeply troubling—legacies. I have gone to the Sacramento Railroad Museum and watched people sobbing about ancestors or kin in the exhibit where they show the history of foreign laborers working themselves to death building the railroad.

If our standard is “never use technology created by organizations that have acted unethically,” then the logical conclusion is difficult to escape. Modern computing, smartphones, social media, cloud infrastructure, consumer electronics, automobiles, aviation, publishing, pharmaceuticals, and much of the Internet all become morally suspect.

Is the expectation that we should reject every technology whose origins or creators have imperfect ethical histories? Or is the expectation that we continue pushing industries toward better practices while evaluating technologies on the value they create today, the harms they cause today, and the direction they’re moving?

The reason I ask isn’t to excuse unethical behavior. It’s because I don’t think LLMs deserve to be treated as though they occupy some uniquely immoral category of technology.

If we’re going to have a conversation about ethics, I think we owe it to ourselves to apply those standards consistently.

If you disagree, the attached checklist in this photo outlines the minimum requirements for claiming the moral high ground.

Thank you for coming to my TED Talk. Please remember to recycle this opinion responsibly.

One of the reasons that there are always so many new ad formats popping up is that our brains identify them as invasive, and we develop an immunity to them over time. Creative advertisers find ways of concocting new signals to assault our sentience, to get at our wallets and perform a Money.

Much in the same way we have set up mental barriers to protect ourselves from advertising, I find that I have erected some communication defenses from AI-generated content.

Let me illustrate that by asking AI to write the opening paragraph for this blog.

I suppose a trigger warning is warranted here.

Trigger warning! AI-generated text:

Advertising has always been an arms race between persuasion and pattern recognition…

A new format appears and works brilliantly for a while. Newspaper ads looked like information. Banner ads looked like navigation. Native advertising looked like journalism. Then audiences learn the pattern, build resistance, and the format loses much of its power.

We may be watching the same thing happen with AI-generated writing.

The first wave of LLM prose felt remarkable: polished, articulate, and seemingly effortless. But exposure creates familiarity… Readers are beginning to recognize the rhythms, structures, and telltale phrases of machine-generated text in much the same way they learned to spot clickbait headlines and sponsored content.

It is not a question of whether AI can write anymore… it is a question of whether readers can tell that it did.

It is not an arms race for generation… it is an arms race for authenticity.

And prompting is rapidly becoming more than getting an AI to write well. It is becoming the art of producing writing that can survive an audience increasingly fluent in the fingerprints of its creation.

I am sorry for shitting directly into your brain. I don’t feel good about doing what I just did, but I think you get the point. There is something about my warm and friendly Szeder-esque opening, and how it differs from the schlip schlop emitted from an LLM that you can spot from the International Space Station.

You might also be momentarily confused… Because generally I am here with the “hooray LLM” flag while sitting at the “I heart LLM” fan club officer table.

Yes, I do love me some LLMs and tools to help make things faster and better. At the same time, you might notice me posting obnoxious jokes about the random cold call attempts I get, often on behalf of LLM tools, on LinkedIn.

Let me help you hold these two apparently opposite ideas in your head at the same time. While I love me some “let’s get this done thirty percent faster” technology, I want that extra time to be spent at a higher-quality level.

I do not want to spend that extra time sifting through someone’s promptables. I have been on the receiving end of documents like this, and I will confess that I am about to give Sam Altman and his point zero one percenters a massive boner. The only way to deal with someone who uses a big-assed old prompt to send you a voluminous document of fewmets and digital offal is to do one of those reverse ju-jitsu things. Take their output from an LLM that has plopped itself into the bowl of your inbox, feed it back into an LLM, and ask for it in summary form; We must fight fire with fire.

I have developed a reasonably good feeling for when something coming my way is authentically written and also when it is “absolutely right!”

Cold calling is not the best thing for an LLM to do, and my inbox is littered with deliberately misunderstood replies as I have attempted to use humor to preserve my sanity from the onslaught. I find that being authentic online is best when you are akshully being authentic, not having an LLM pull a “how do you do, my fellow kids?”

Sam Altman ain’t getting paid for making fetch happen. LLMs and cold calling go together like peanut butter and dynamite.

I am not going to hop on a soapbox and attempt the moral high ground argument around the environment, or the cost, or the … whatever it is people cry about as they tattoo “FIRE MY ASS” on their foreheads as they pop off on LLM technology.

I will say “Yes, this is one thing it does poorly” and then move on like an adult, capable of agreeing to disagree on the internet’s vast ocean of (mostly wrong) opinions. I won’t scream “SOURCE?!” in your face or hold up your arguments and counter assault with the fifth grade’s best ad hominem attacks.

I mean… “I know you are, but what am I?” In 2026? We can do better.

All I want to do is make sure you know that some people love communicating with other people. And I am one of those people. Attempting to impress me by a “save as” PDF from Markdown with Claude colors draping the document is not going to get very far with me if you want to impress me or do a Money from me. Just no. Stop.

At the very least, I want you all to know these people are out there, and it is okay to take what they send you and feed it into an LLM to ask “what means?” or equally toss it in the garbage and pick up the phone.

That is it. That is the whole post.

Speak to you all again soon.

I am not a popular person on LinkedIn with the “LLM are bad” crowd. Most of my friends are in this unpopular group, since we have all just decided that we would rather explore the potential of these tools, rather than pretend to spit in the direction of their headquarters. You can find these conversations if you really want to. I am not going to promote them. I have been tagged in posts by enough irrational haters for one weekend.

Make no mistake. I have yet to have many healthy conversations with people who have the “LLM are bad” merch proudly sewn into their shirt or stapled to their forehead.

If you believe “LLM are NOT bad”, be prepared for some angery words. I am an enthusiast for the artistic output of next-generation creators using LLM tools, and heartily cheer their work.

I am not ashamed to admit this. I will say that it has gotten me blocked by some people on LinkedIn, tagged in threads by people with very mean words next to it, and generally shocked at the sheer level of toxicity in people’s opinions.

I have even had to start deleting things from threads on my page and putting in a warning that I consider the boundary of Godwin’s law to be the line I draw comments at. You can Google that if you don’t know what that means.

It is absurd and crazy to me that no one wants to accept an “agree to disagree” position here; for some, it has to be all or nothing.

Generally, after it is clear that I am not moved by their tantrums, petty name-calling, and emotional foofaraw, they drop the following bomb.

“You will be sorry when the bubble bursts.”

I want to unpack that a little as I rant about the future of AI and the future of games.

For starters, I think it might come as a shock to some of the folks out there that this is not society’s first tech bubble.

There was a web3 bubble, a mobile 1.0 bubble, a casual games bubble, a social platform bubble, the dotcom bubble, and, for that matter, the “Atari ET buried in the landfill bubble.”

I have lived and worked through most of these. And ultimately, while some percentage of investors with actual dollars took a bath in some of these bubbles, we all managed to survive, and in some cases, we continued doing what we were doing.

Yes, there will probably be an AI bubble. We are going to see a lot of people lose investment money in the frothy run-up to this bubble, and it might be happening RIGHT NOW.

That does not mean AI is going to go away. Even if Anthropic, Claude, and all these big frontier model companies stopped making new models today, I would still be running a local model that does some of the same things their current models do, maybe 25% slower or 25% worse. The gains from using LLM tools are there. I wish I remembered the study link where someone provided proof that 80% of all business activity that people do is better done by AI.

If you find that study, please drop me a note. I would love to have it handy when angry, flustered meme-able people stick their face in mine and scream “SOURCE?!”

It happens quite a lot.

It does say to me that we are going to be seeing a massive change in the future of work. There are whole industries of middle-tier people whose job right now is to take some level of information, manipulate it or synthesize it, and then regurgitate a report to some high muckety-muck. In the process of doing so, they can make recommendations, suggestions, and improvements that will get them promoted.

The last part of that sentence, the recommendations, suggestions, and improvements? Those items will include some level of easy, low-hanging fruit that can be generated by an LLM. Some of it will be stuff that requires what I have called “Platinum Collar Thinking,” which is a fancy way to describe the remaining 20% of work that is best done by humans.

That will still exist, but people with Platinum Collar Thinking can’t just sit around and do the 20% of the work that used to exist and expect to be rewarded 100% for it. They are going to have to do four to five times as much work as they did before, in ways that matter to their employers.

We are all going to have to get much better at doing our everyday things when we drop some information into the Claude-inator 2000 and have it emit some CSV or a report that might need a little finessing or “interpretation.”

So let’s bring that to games.

We are seeing new inventions, new tools, and new technologies leveraging LLMs showing up every day.

We are not yet in the territory of having our progress bounded by stuff like Moore’s law. I have twice run into technology-based companies that are doing two or three orders of magnitude improvements over existing LLM capacity, quality, or speed. Some of these improvements are even stackable.

While these advances are largely bleeding-edge and are only now entering the wide, wide world of people who are “level 1” on the four levels of AI adoption, we are going to see huge transformative changes in the way we think and the way we work. I attended an event called “FrAIday” here in my town, where we had a deep hour-long conversation about this.

But the biggest thing that is going to change?

How we play.

This is really why I get called all kinds of names and people block me for my “AI-First” position.

We have been through multiple waves of technology change in the past thirty years. I have had a front-seat view of much of it.

The internet. The first smartphones. The social internet. The touchscreen smartphone (mobile 2.0).

I also meandered through a few near misses. Virtual reality. Web3. The Metaverse. All three of these will manifest themselves in coming waves of technology and how we interact with machines, and each other, and how our machines interact with other people’s machines.

The wedge in previous iterations of technology, going all the way back to before my time and including things like the printing press and television, has always been entertainment.

And the same thing will be true here with LLM tools.

And this is what is generating all the fear and confusion, and people throwing tantrums.

We are going to see new genres and new forms of entertainment using LLM technology coming fast. We are going through most of the last waves of “just use LLMs to supplement parts of the pipeline to make existing things.”

Consumers are not excited by that, and the industry does not love it because it is already in a massive down cycle for other reasons. If you spend less than five minutes thinking about the art that is possible with LLM technology, it is really easy to get a low-level threat on Maslow’s Hierarchy of Needs and react with “Oh Oh, It Is Coming For My Jerb.”

This is especially true if you have already lost your job. But right now, that job isn’t coming back anytime soon. And this is a convenient scapegoat for it.

The pattern I see in some people with a lot of grit and a lot of curiosity and a lot of the people I have been fortunate to work with is to lean in, learn something new, and continue to make things. This is true for people I have worked with for one year, and this is true of some of the people I have known and appreciated their professional partnership on things for thirty years.

I spooked the living shit out of a senior creative director, showing off my vision for the future of entertainment. “In the world you are describing,” he said slowly, “The job I have right now goes away.”

Yes, it does. But a dozen new jobs will arise in its place.

Much like when the printing press arrived, and the ownership of books and the ability to learn things and get educated transformed with the lowering of the barrier of cost of creating books, we all benefited, and society moved forward.

You see lots of big celebrities and senior creative people talking about their concerns. You are damned right. They have been gatekeepers for professional success in the world of leisure. And those gates are being blasted open. The ability to create something will no longer require up to nine thousand highly paid people to generate something that is worth consuming.

Solo creators and smaller teams will throw their creations together using well-articulated designs, LLM tools in partnership with niche craftsmen and tastemakers, to redefine the very things we do for fun.

The problem with this is probably one of timing. The hits to the creative industry caused by interest rates, changes in consumption patterns caused by COVID, and eventually the “end” of COVID, and other factors for games like the growth of Roblox, and in Hollywood, things like the writer’s strike and the inability for studios to take serious risks on fresh approaches to content have created a window of opportunity for new things to emerge.

In the next five years, we are going to see a further decrease in the number of AAA titles that are shipped. We are witnessing the slow meltdown of the console space, where everyone has raised their hardware prices due to the component demands, which will complicate next-generation designs and launches.

We have never seen such a radical transformation from top to bottom of just about everything.

I have spent most of my thirty-year career doing zero-to-one startups and dealing with crazy levels of uncertainty. I was built for this kind of low-oxygen experimentation, and I love the challenge of trying to understand how to thrive when this level of uncertainty exists.

I also think this is probably the biggest technological change we are going to see for a while, and there are only going to be two or three waves as big, or bigger, possibly ever.

So why am I writing this?

Because if you have your head in the sand right now, you are going to miss the opportunity of a lifetime.

I really don’t care about all of the name-calling, blocking, and gaslighting I have received simply for having this opinion. I don’t care for it either, honestly.

But I want to explain why I am excited for the opportunities that are unfolding and why I always lean into what is possible.

I am curious to see how this all shakes out by 2030.

How about you?

If there is one thing I love, it is a mixed metaphor. 

“He climbed a ladder of stability across an ocean of trouble and left footprints in the face of time.”

This glorious quote comes to us from an English writing book we used in High School, three, or five, or mumble mumble years ago. The exact year does not matter. We will move on.

It is relevant because of where we are in the world and the current timeline we are on, if you believe such things.

Today, I want to talk about the dream of The Long Tail, and more specifically, about how it is turning into the Emperor’s New Clothes.

I think that probably gives you an idea where I am going with this.

You can all go to the Wikipedia or the Google and do a history of the Long Tail. The idea is that everything keeps making money over time because the internet is forever.

While there is some truth to this, the Long Tail also has a diminishing cost to it. Eventually, it gets to the point that it is below the threshold of adding it to a balance sheet (under a penny), and some time before that, it is below the threshold of adding it to a payout (most places are 10 dollars or 100 dollars to consolidate transaction fees).

The Long Tail at that point becomes meaningless and invisible, and suddenly a joke, which is what brings us to the Emperor’s New Clothes. You can look up that story if you need to, too. Essentially, it means we were sold this miraculous world where everything keeps making money until the heat death of the universe. It is simply untrue. At some point, the content revenues fall to the point where it is not cost-effective to maintain the payment relationship, and the whole thing atrophies.

This is how we have gotten the Emperor’s New Tail, in case you were not following.

So what can we do? I thought the future was going to be abundant for everyone!

Yes and no.

The good news is that having everything available everywhere, all the time, means that at some point, a person of influence is going to talk about your stuff, or perhaps improvise a dance to it.

I present to you the song that creeps out my wife when I hum it.

It is a great tune from a great artist, and it happens to be catchy. It also happened to be used in a very creepy movie my wife regrets agreeing to watch.

TL;DR, if you are on a hiking date and you see this on your hiking partner’s playlist: Run.

Back to the E.N.T. 

Everything needs to be available everywhere, even if it is sitting in the middle of the Island of Misfit Toys. Eventually, someone is going to make Fetch happen, much like “Go” was brought back into the eye of consumer-Sauron for another fifteen minutes of fame.

You will need to accept that the Long Tail is not a forever Tail, unless you have tattooed it to the face of your children, or laser-etched it to the face of the moon.

I wanted to talk about this for a few reasons.

First: If you are a game developer and you are relying on Steam to save you from being poor, you might need to spend a little more skull sweat contemplating your Make Money Fast plan. There is no place where the Emperor’s New Tail is more clearly real than the Steam catalog. Okay, sorry, that might apply to Spotify. Or Goodreads. Suffice it to say, there are lots of places where the pennies per unit creativity extraction is very bad and getting worse by the minute. If you intend to play in this playground, bring serious marketing plans, or you won’t be there for very long, if at all.

I am going to spare you the “Ladies and gentlemen, let’s talk about AI!” circus pitch, and why this matters. The people who need to hear the E.N.T. story are largely in the “you cannot spell horrible without AI” even though AI is not in the word horrible. I will exercise an Empathy card here and give you a pass on that subject because you just spent your last 100 dollars submitting a potential game of the year to Steam.

It doesn’t matter that you are not likely to make that money back. You bought five seconds of life for Gabe’s super-yacht, and because you are a fanboi who loves him and his 30 percent tax on your moneys after 100 dollars, that is all that matters.

The second reason I want to talk about this is that we are on the cusp of a content renaissance. At no point in history has it been easier, cheaper, or faster to create stuff that other people will want to stare at and listen to.

We are approaching a period in time where the size of the audience of consumers is going to be in the same order of magnitude as the size of the audience of creators. The walls are falling, and the ability to gate-keep this much content is nigh impossible.

We are drifting into unfamiliar space here. The rate at which stuff is changing is going to keep at a frenetic pace for the next few years.

While we want to approach the future with cautious optimism and a hope for abundance, it is also a good idea to keep the shields at maximum.

And maybe have the crew on yellow alert.

I appreciate you being patient for this post.

Thank you, as always, for your patience.

See you next week.

Get outside the inside of your head

The reader’s choice topic month came and went. Thank you to all who voted. It wasn’t like there were thousands of you, but it was enough to help me break through a “what the hell do I post” deadlock. Onwards and upwards.

This week, I wanted to armchair a little on the world we live in and spot the best places to deploy AI. Where are the “aipportunities,” shall we say.

There are two ways to go about this.

The first way is to look for blue oceans and greenfield opportunities. Where are the new categories springing up, and where are there places for things to exist where they never existed before?

This is your agentic workflows, your vibe-coding tools, and your openclaw instances hard at work.

You can GPT up your Chat if you need to know more, or attend a meetup.

The other really good way is to find someone who has set themselves as an anti-ai platform and is viciously spending time and money defending their moat.

There are three obvious leaders here I want to talk about.

The first is LinkedIn.

The second is (shocked Pikachu face) the games industry.

The third is Upwork.

Let’s talk about these, shall we?

LinkedIn has set itself up to be a laughingstock as a platform. If you have not seen one of my “It is Day NNN” posts, you should creep me over there. You might recognize what I am posting about because it is infested with horrible noise.

The noise is largely because the platform has done its best to keep collaboration with AI to a near minimum, and I have heard horror stories about people being banned from LinkedIn for experimenting with API tools.

Considering they are owned by Microsoft, and seeing literally every misstep possible being made by Microsoft, this is just another leadership category error in the making.

After LinkedIn, there is “the whole games industry.” Games have 100% always been at the forefront of every technological wave of innovation. Except for LLM technology. It is borderline lunacy how toxic LLM technology and LLM developers are treated in games. I am on the receiving end of this myself, and I have not even shipped an LLM game yet. I get people sneering at me, making horrible comparisons in passive-aggressive vaguebooking posts, and flat out being blocked for being enthusiastic about the incoming change in tools that we will be experiencing. I do my fair share of trying to help people agree to be open to new tools and processes using LLM tools, but I have stopped being kind after a first-degree connection literally was gaslighting me and attempting to patronize me into accepting that his toxic treatment of my opinions was normal and that somehow I am more stupider for not understanding why I am so horribly bad.

I am going to painfully wait for the tantrums and thumbsucking on this front ride itself out. There is no other real alternative. I am not going to attempt gentle Socratic conversation anymore. I am just going to point out they are throwing tantrums and point out that “The Big Beige Book of 2028 Employability” will require people who can spell LLM. I describe it as painful because any time I try to offer a constructive path forward for people, they smack it down like a toddler breaking a plate full of delicious vegetables. Then they scream in my face, “NO LIKEY!”

At some point, like every other technology wave before this one, someone is going to ship a hit game, and then ten other companies are going to knock it off. Then AI Native games will have a proper genre, and we can all go back to normal.

That brings us to Upwork and its ilk. Much like LinkedIn, here is a protectionist ratings racket for work. They have APIs that they apparently hoard zealously in a pathetic attempt to extend their moat for pushing developers into low-leverage roles, where customers can threaten them for discounts due to one-sided reputation management. “Give me an extra milestone’s worth of work or you get one star,” is a motivator for someone who is already struggling to make ends meet. That lack of reputation will make it harder for you to command your existing clientele or wages.

I have more on how Upwork will eventually perish in the darkness where it thrives, but there will be trustable transparent platforms leveraging LLM technology to make better developer pairings with projects, and also whole new ways to do this business without having it be a complete race to the bottom that commoditizes hard-working creative people. You might have to wait until mid 2027 to hear more of this. For now, I will keep it secret, I will keep it safe.

So here are three places you can make a tectonic impact against competitors playing “scared ostrich” for their strategy. Pick one of these, or find something that has a similar set of wrong-thinking leadership forcing people into business relationships or using scarcity tactics.

There are many others out there. If you pick one, figure out a transparent model, and figure out how to create value and joy for people through the wonder of next generation LLM tools, you might be able to take over a shameful protectionist who is huddling under their desk day drinking and wishing all of these new developments would stop so they could just croon lovingly with their fax machine and their horse and buggy like the good old days.

Until then, bad man Darth Szeder will go away for a little bit so you can put on some Kenny G, take out the tub of ice cream from the freezer, and make some chamomile tea, or whatever it is you need to do after you absorb his mean old words.

I will undoubtedly be back for more of this next week.

Maybe with an Amazon Affiliate Link. You know, because apparently I want to push more money at mister Bezos, since I am already not a bad enough mans.

It is somewhat funny that everyone is now talking about the cost of AI. Over two months ago, I decided this was going to be a topic of conversation I wanted to address, and I am not even counting the toxenmaxxing shenanigans, which are ultimately noise.

I have now had about six noteworthy LLM-assisted customer service calls. There might have been a few more that I have not paid close attention to. I am going to talk about two of them in particular.

First when I called a really fancy local restaurant to let them know we were going to be about 15 minutes late getting to our reservation due to delays getting the kids to bed before dinner.

The first noteworthy thing about the call was that it was clearly an AI voice. They made it cute by adding a little background noise, including some typing noises while looking up my reservation.

When I explained the issue, that we wanted to inform the hostess that we were going to be late, it took a few different tries to explain “we are on our way but will be fifteen minutes late” before the agent understood that meant we wanted to speak to the hostess. It struggled with the idea that we were going to be late, but it did understand that it was a call worth escalating.

I waited for a moment before the call ended. The “forwarding to a human” part did not work correctly, and the call was dropped.

This brings to mind the second time, as I was onboarding to a payment provider as a contractor for one of my customers. This is a fairly prolific website, and I wanted to upload my EIN information and details about who I am and what I do.

Their fill-out-the-pdf page was busted, and I got escalated to email by a bot, which confirmed that I had a bug.

The email from the payment provider insisted it was a real person and they wanted to gather more details about the issue, including traces from console output, HTML, etc.

I provided all of this, and two times later that day went to test something that still didn’t work.

This standoff lasted about an hour before I decided to just not care anymore until one business day had elapsed.

The good news is that on day two, the Edge browser worked for me to solve the problem of giving them tax information from the browser, and I could go about my business.

In both cases, the AI agent did not do much to solve the problem, nor did it engage meaningfully.

The failure to do the job correctly for both of these AI implementations, and the damage that it does to the customer relationship, is the cost of AI I want to talk about.

Sure, $500M in tokens is expensive. Companies have spent as much or more trying to rescue brands in the past. In the case of a fancy restaurant and a fairly large billing portal for vendors, these are places that can ill afford to alienate their audience or break trust. Breaking trust is exactly what happens when you deploy an obvious AI customer service system that fails to do its job correctly.

And this is where the real cost of AI will come into play.

There are some things that AI is really good at.

There are some things that AI is not really good at.

In fact, it can be outright terrible, especially if it is easily breakable.

I think everyone has been in a mad rush to show they are using AI to do something, and they went overboard.

Any time you are putting your brand in front of someone and backing it up with an AI resource, you should be really sure it works and that it does what it says.

I met a founder who is making AI tools for the customer service space, using them as training tools to help with role-playing for agent training. Otherwise, people are doing mock exercises that come across as insincere, or they are learning in the crucible of fire of live customer experience. Neither of which is great. The AI tool records the conversation and finds weaknesses and strengths for the future, and maybe helps with a remediation plan for weaknesses. This is a good use for AI in customer service because it only helps the customer.

There might come a time when this will be laughed at as a naive and old-fashioned story. I think that time is probably four to five years away. We will see more and more agents out there doing things, but if you just bolt an agent onto a process and it is not sufficiently baked, you are going to do harm to your service or product.

I am writing this because I spend a lot of time telling people what AI is great for.

I think we should also spend time talking about what AI is not great for.

So if you are about to embark on an AI project, ask yourself some questions about who is going to interact with it, and then, if they have a broken experience, what is the price of that interaction?

The companies that navigate this conversation the best will be the ones with the most valuable brands in the eyes of the consumer. Everyone else will find their business slowly bleeding out and not understanding precisely why.