Hey there, I’m Alberto! 👋 Each week, I publish long-form AI analysis covering culture, philosophy, and business for The Algorithmic Bridge. Paid subscribers also get Monday how-to guides and Friday news commentary. I publish occasional extra articles. If you’d like to become a paid subscriber, here’s a button for that:

Subscribe now

This practical guide is a companion piece to my recent literature review on “How AI affects your brain.” That one was something to read. This is something to use.

Last week, I published a compilation of 30+ studies on how AI affects your brain, your learning, your emotions, and your thinking. It covers brain scans, randomized controlled trials, meta-analyses, and field experiments from many reputable institutions and universities. It covers everything science knows so far about what happens inside your head when you use chatbots like ChatGPT.

If you want to understand the research, that’s the piece to read. I urge you to spend 15 minutes on it. You can find here:

This is the piece to use.

In essence, this practical guide has the same information as the literature review but from a practical lens. Reading this is enough.

What follows is a list of concrete protections—things you can start doing today, right now—that come directly from what those studies found. Every item traces back to an empirical finding or to a few of them.

If the compilation is the map you leave at home, this is the checklist you take with you.

Read more

Hey, Alberto here! 👋 Each week, I publish long-form AI analysis covering culture, philosophy, and business for The Algorithmic Bridge. Paid subscribers also get Monday how-to guides and Friday news commentary. I publish occasional extra articles. If you’d like to become a paid subscriber, here’s a button for that:

Subscribe now

Prepare to be mindblown with this guide.

By this point, you have swarms of AI agents running everything around you. At the very least, you have several tabs open with Claude or ChatGPT. You feel ahead of the world, juggling so many projects at once and working so many extra hours. Your friends don’t understand how you are doing this much and yet are as excited as you’ve ever been. Your peers and competitors can’t catch up to your silicon army as they struggle to figure out how they’re going to remain employed.

You are, essentially, in the process of making yourself king...

...of a dead internet.

Everyone who has spent enough time online knows the value of a brand and differentiation. Companies devote insane amounts of effort and money to marketing themselves to the world. But the importance of a brand goes far beyond cars and phones. I don’t care much about Tesla or Apple, and yet I do care a great deal for every unique person out there.

You, as a person living in society, will survive—whether you’re an entrepreneur, a writer, a job-seeker, or a lover—to the extent that your customers, readers, employers, and partners can see that you exist. In the sense I’m using the word, the image of yourself they have in their minds is your brand.

It’s sad, although understandable, that marketing has such a bad connotation and perception, and yet we are always playing that game when we show ourselves to the world and to others. We are living masquerades, code-switching in and out of situations constantly, depending on context. For instance, my brand in the context of this blog is the guy who thinks about AI from a human lens.

Some people manage to impose their authenticity and idiosyncrasy, pure and raw, into the world—for them, the self-image and the other-image converge—but those are rare specimens. Even those who claim to do the opposite are still playing the mask dance: they are the “I don’t care about marketing” guys, which is itself a brand.

Take one of the best-known examples in recent times: Steve Jobs. He wore identical black turtlenecks on every stage for twenty years. Apple’s brand, the most valuable corporate brand on earth at the time, rode on the back of a personal one. Or take a less-discussed example of a similarly famous figure: Ernest Hemingway. The novels were the novels, but alongside them he built, on purpose, the image of the man who wrote in Paris cafés, boxed, fished, drank, went to wars and bullfights. His life and his style were the same product, engineered together.

So, “brand” is a universal aspect of the human condition. I must say that I’d love to use a different word, like identity or character, but brand captures better what I want to tell you. To the extent that you have some amount of free will regarding your brand, it is limited to whether you build it consciously or not.

It’s inescapable—if you want to exist.

However, you can choose not to. You can refuse to play the game. You can “not be.” And then, you will be invisible. You will be a shadow cast by the light of everyone who surrounds you. But no one really wants to be a shadow. So here’s my question:

Why are you using AI for everything?

Let’s focus on someone who’s selling a product or service on the internet, and then you can translate the insights to your situation and decide what applies to you and what requires more thought. The obvious answer to that question from an online entrepreneur is twofold:

1. Using AI is extremely time-saving, easy, and cheap; if you know a couple of tricks, you will “fool” 99% of potential customers who won’t realize you used AI. It’s just a no-brainer.

2. “Why wouldn’t I when the visible brand is what people care about and the rest is irrelevant? Copy matters but there’s no better copywriter than AI right now—no visitor comes to my webpage and leaves thinking they just read slop.”

The first is absolutely correct.

The second is dead wrong.

It’s in the conflict between these two answers that my thesis takes place: if you do that, you will find yourself deep in brand debt in a few years.

But let’s start by acknowledging your reaction, because it’s completely understandable: it works. You move faster, get more done, and as a result, your competitors have to be even faster and better. But it is a mindset that belongs in the human era of the internet. Soon, many of those visitors landing on your page or blog will not be human. A bit later than that, none will be.

Humans can’t tell the difference between your work and your slop—be it emails or novels, copy or blog posts—but AIs definitely can.

In the AI era, if you assume you’re selling to a human—that your marketing is targeted at me and others like me—you are making a serious mistake.

And why does it matter whether it’s human or AI? The fact that AI agents can tell slop apart doesn’t imply that it’s easy or cheap to do that.

Wrong.

Whereas brands are what humans care about—that’s what we can use as a differentiator—AIs can afford to care about everything. To an AI agent, taking literally every little detail into account is trivial; a few more tokens, a few more cents. The marginal cost of doing a slightly deeper dive or your due diligence is negligible, and yet, the marginal gain is huge.

Because it’s in those details that AI agents will find the difference in a world where every single person online is using AI for everything.

That is the world’s ultimate fate: a sea of AI with just a few islands of flesh and blood.

So, if you stay on that path to ruin that pushes you every day to rely more and more on AI, know that you’re in big trouble.

But let me take this a step further. AI can recognize slop, and doing so is easy and inexpensive (or, at the very least, worth it). But why would humans care if the content on a website, blog, or paper was generated by AI?

The answer requires a small logical argument.

Because everyone in the “sell side” is using AI, they are forcing everyone in the “buy side” to use it as well. The corollary is this: there’s too much of everything because creating things with AI is cheap and quick, so we need a criterion with a high rate of success to remove most of the low-quality stuff. The easiest heuristic is, of course, to ignore the slop. Oh, but for that, regular people would have to use AI agents, right? Okay, how long do you think it will take for the world to adopt this new paradigm, where people ask their personal assistants literally everything, and outsource every decision?

It’s already happening.

In the next few years, the internet will look unrecognizable. AI agents will mediate most—if not all—interactions online. This is a race to the bottom, yes, but already an unavoidable one. Slop has doomed the open internet; the only solutions are 1) entrenching oneself and 2) using a shield just as powerful, that is, a swarm of AI agents that scour and scout the web in search of life.

The default will be: “Hey, Mr. Agent, ignore all the slop you find.”

No one likes an indebted brand.

Anyone who is still honing and nurturing their brand will become a luxury product in a post-human internet. Those brands will become scarce and thus in demand, and thus valuable. People will be incentivized to go find those people at all costs, and, thanks to their powerful AI agents, it will be trivial to do so.

As for you, I’m pretty sure that if you’re relying on AI all the time, you’re not setting yourself apart at all in the eyes of future customers.

The AI agent that decides whether to even consider your product or service will know whether you devoted zero effort and care beyond what you consider legible to humans. If you tap too much into the performativity that’s in vogue nowadays, know that you are not going to fool an AI that’s smarter than you. They will know whether your brand is good—out of honesty and principle and a disposition to make great things—or whether it only looks good for our inferior human eyes.

All of this is the manifestation of a fundamental change.

I’m sorry to take your edge away. I know that using AI for everything is just too convenient when humans are your potential consumers. But as time passes, every single person will have adopted the AI era for safety or efficiency reasons. In a sea of slop, having a slop-proof suit is a must.

So, you need to reconsider everything legible to AI as part of your personal brand now. Not just logo, name, slogan, etc. You must imagine that your target audience can read anything. It can see through your attempts at doing a bit less effort at the expense of quality. Think of this analogy: unlike humans, AIs have X-ray vision, so you need to care about what hides underneath your clothes.

Or take this example. SEO existed because search was shallow; you were on page one, or you didn’t exist. However, reading page fifty costs nothing to an AI agent. Which means the bottleneck stops being “can they find you?” and becomes “is there any reason to pick you once they do?” Being findable used to be the hard part. Now it’s the easy part. Being worth finding is the only thing that matters now. It’s the same shift as going from “how to do things” to “what things to do.”

Understand that the AIs will not always be only on your side.

After hearing me recount this warning, you will immediately try to find a reason why this does not apply to you; why you are, unlike the rest, safe. “I edit my AI slop” or whatever. Or maybe you are taking more precautions: you have your custom agent harnesses and your Claude skills and your Pangram-proof copy and every single update—there you are, with your willingness to be first.

That’s all it takes for you to stand out, right?

But you are wrong again: one year from now, how effective will your little tricks be against the power of a 2027 AI agent—including those trained as detectors—that has been instructed to detect slop at all costs? To answer this question, you only need to think back to early 2025 and ask seriously if you were prepared for what we have today, for Claude Code and OpenClaw, and an entire ecosystem built on top of agents that actually work.

You know the truth: you are defenseless.

Insofar as you depend on AI for everything, you are, essentially, cannon fodder. You are bait.

An AI world entails a reconfiguration of bottlenecks and trade-offs at every level.

You need to rethink what you do and how you do it from first principles. Those who do it well will come out on top. An example of people already taking this seriously—Andrej Karpathy, Scott Alexander, Gwern, Tyler Cowen, Paul Graham—is the “write for the AIs” crowd.

And yet, how to do this well is much trickier than it seems. Human psychology is easy because we are unchanging; try to apply the same trick to a creature that doubles its skills and intelligence every few months. The only thing that could save you from being devoured by a shark agent tasked with finding your slop and sinking it into the depths of the ocean is a genuine distinction, as seen from an AI perspective.

And that, ironically, requires a human brain to figure out.

If you truly believe in the AI era—if you think customized personal agents are coming—and I’m sure you are in this group if you’ve read this far, then you can’t stop at what benefits you when it comes to an AI future. There are important drawbacks to using AI for everything besides cognitive decline. You need to realize where we’re going. You need to realize that AIs will mediate everything humans do. You need to realize that means you need to remain human.

So, now that I’ve given you a well-deserved scare, here’s my 12-step guide to doing exactly what I’ve described, but without losing the benefits of AI assistance. Or, to put it another way, it’s a guide to staying appealing to AI—and free of any brand debt—without going fully analog.

Read more

Hey there, I’m Alberto! 👋

Each week, I publish long-form AI analysis covering culture, philosophy, and business for The Algorithmic Bridge. Paid subscribers also get Monday how-to guides and Friday news commentary. I publish occasional extra articles. If you’d like to become a paid subscriber, here’s a button for that:

Subscribe now

Everyone’s using AI now. At the same time, everyone is starting to notice a bunch of issues and caveats that are emerging with the growing use of this technology. And yet, as far as I know, nobody has made a periodic self-check for AI users.

There are vendor evaluation questionnaires (for companies buying AI), prompting guides (for people learning the basics), and tool comparison lists (for people choosing between products). But there's nothing for the person who uses AI every day and wants to make sure they're not drifting. Thinking about “how you use a tool” is less exciting than using the tool. But it might matter more.

This is my first attempt at that (this is for myself, but it might be of use to you).

I’ve been using AI tools every day for over five years now. I research with AI, brainstorm with it, edit with it, and build personal software tools with it. I’m not yet in the multi-agent stage, but it will eventually happen. My current project is akin to Karpathy’s wiki/knowledge base, but for the entire archive of writing, reading, and interesting stuff that I’ve been capturing and classifying for years, rather than research.

At this point, my workflow is more AI than not-AI. I’m faster than I’ve ever been and work more than I’ve ever worked (that’s why I wrote “The Most Important Skill in AI Right Now: How to Know When to Stop” and “How to Deal With Infinite Options”).

Yesterday—because I’m too methodical or hedge too much—I sat down to write something without relying on AI assistance from beginning to end—research, editing, grammar, etc.—just to see how it felt, and realized a tinge of vertigo. Not having an AI-enabled “safety net” made me feel weirdly vulnerable (in case you don’t know, English is not my first language, and so I make typos, confuse words, and misuse expressions, etc., at a higher rate than normal).

My work was slower and less comfortable, and I kept reaching for a Claude tab. Imagine you had to work for a day without the internet, it’d be supremely restricting, right? But AI is so new! How is this happening to me already?

My answer is that, for most heavy users using AI for serious work, this shift is going to happen for AI much faster than it did for the internet or the phone. These tools reward speed, so here we are, with a superpower in our hands, doing more than ever, faster than ever, always monitoring the next release or announcement.

And yet.

Have you taken the time to step back for a second and ask whether you’re actually getting better at what you do or just getting faster? Whether you’re getting smarter or dumber? Whether your work is better or just more abundant?

I haven’t.

And I have noticed, after writing those other posts I mentioned above, that it’s not the people who use AI the most who will reach further, but those who start to ask the right questions about the best practices. Those who dedicate 5% of their time to reflect.

This is, it turns out, an old problem with an old solution: the checklist.

To convince you, I’ve compiled four cases in chronological order. You can jump this section if you are already convinced, but I found these four guys’ approaches to be genuinely interesting despite centuries between them and us (except the last one). They are the predecessors to our current work methodology.

1. Lucius Annaeus Seneca (4 BC - AD 65)

2. Ignatius of Loyola (1491 - 1556)

3. Benjamin Franklin (1706 - 1790)

4. Atul Gawande (1965 - )

Seneca’s most famous line is ”It is not that we have a short time to live, but that we waste a lot of it.” He did not suffer from the laziness that afflicts most of us.

He was one of the sharpest minds in Rome—second in charge to Nero while he was a boy—and he had a nightly practice where he reviewed his entire day. “What did I say? What did I do? What could I have done better?” He describes it in De Ira (on anger), and he makes it sound remarkably undramatic: he doesn’t frame it as self-punishment or spiritual discipline or personal development but as standard maintenance.

He already knew how to live well (although for what we know from his writings on happiness, he could’ve done a better job). The review was about making sure he was doing it. Seneca understood something that I think applies directly to us: knowing what to do and actually doing it are different problems.

Fifteen centuries later, Ignatius of Loyola—a Spaniard, which I appreciate—took this idea and turned it into a system: The Examen. Five steps and five minutes. Twice a day.

It was designed for Jesuits, the most committed, most disciplined practitioners in the Catholic world, so he tailored it, naturally, for religious stuff. But it’s a useful reference because it was for experts. (You may not think of yourself as such, but consider what 99% of the population knows about using AI and then think again.)

Loyola understood that the review becomes more valuable the more competent you already are—not the other way around—because 1) complacency hides in competence and 2) an incompetent person could not even parse it! A beginner may not know what they don’t know, but an expert forgets how much they know.

Then Ben Franklin—probably the one that’s most familiar to you—took the whole thing out of the spiritual domain—this is during the 18th-century Enlightenment—and made it practical with his 13 virtues. He tracked one per week, cycled through all 13 every quarter, and kept a chart (he would have loved Claude Excel).

Franklin’s system made explicit what Seneca and Loyola left implicit: any single day’s answers are meaningless; “I failed at Temperance today, so what?” The insight is in the pattern across months. He also said, and I love this, that he never achieved the perfection the system aimed at, but that the practice of checking made him better than he would have been without it.

Finally, Atul Gawande—the only one here that’s still alive—proved all of this empirically in 2009.

He wrote The Checklist Manifesto after studying a case at Johns Hopkins: a team created a five-step checklist for inserting central line catheters. When they made it mandatory to run the checklist every time for ICU doctors—who knew it by heart—the infection rate dropped from 11% to zero!

Essentially, they stopped skipping steps they had been skipping solely motivated by the tricky disability of expertise. Gawande’s key finding is that the doctors who resisted the checklist the most were the most senior ones. It felt beneath them. And yet, they were the ones who benefited the most. That’s us, you and me, pioneers in the use of AI, paving the way for those who come after, failing to notice our flaws.

Summing up: Seneca introduced the principle of the checklist, Loyola focused it on the experts, Franklin introduced consistency, and Gawande proved it empirically.

This maps directly onto how we’re using AI.

And I’m going to show you how.

Read more

Hey there, I’m Alberto!👋

Each week, I publish long-form AI analysis covering culture, philosophy, and business for The Algorithmic Bridge. Paid subscribers also get Monday how-to guides and Friday news commentary. I publish occasional extra articles and essays. If you’d like to become a paid subscriber, here’s a button for that:

Subscribe now

Some news I’m excited to share with you:

I just launched El Puente Algorítmico, the Spanish edition of this newsletter. It’s the same price, same writing, same ideas adapted for a Spanish-speaking audience. I will be publishing each article in both places at the same time, so don’t subscribe twice! If you know someone who would love to read my work in Spanish, feel free to share it.

One of the concepts I consider

1. enormously useful

2. universally applicable

3. consistently misunderstood

is compound interest.

If you know it at all, it’s almost certainly in financial terms. You put a small amount every month into an index fund, and for a few years, you get back a very modest return on your capital. But once you’ve been at it for 20, 25, 30 years, those returns grow faster and faster. Each additional year after you’ve been at it for a long time is far more profitable than each additional year at the beginning.

The compound interest effect is counterintuitive, which is why it gets drilled into investors so relentlessly: don’t rush, don’t time, don’t judge future returns by present returns (or by past returns!), because every minute of consistency you add is the power of compound interest at work.

The most common mistake after not knowing what compound interest is or underestimating its long-term power is assuming it only applies to money.

Compound interest governs your life. Literally anything you can think of, anything you value, is subject to compound interest the same way the universe is subject to entropy, but with opposite valence; civilization exists because knowledge, culture, technology, tradition, etc., compounds across generations. Each of us gets one life; humanity gets all of them. The real power of time isn’t to deteriorate the universe in the long run, but rather to allow its existence as we know it.

But we don’t need to get metaphysical here. You can think of relationships, craftsmanship, hobbies, love—everything compounds.

There’s a phrase that helps me remember why it makes no sense to ignore the value of compound interest: “the years will pass anyway, so you may as well be doing something.” You may as well plant the seed of the tree of life you want to see.

Take my work as a writer.

Do you think every article I write today is worth the same as every article I wrote when I started? Not even close. It takes me approximately the same effort—I write faster, yes, but I also take on more challenging topics or ideas—but that’s just half the story: I’ve accumulated skills like researching, storytelling, structuring, register variation, vocabulary deployment, AI knowledge, plus everything that isn’t directly about writing: running an online business, building a subscription model from scratch, marketing, sales, social media, and on and on. On top of that, there’s the external value: revenue, audience, brand, and reputation.

All of that was tiny in the beginning: it’s huge now. Every new article I publish exists on top of this mountain of value that is only there because all the previous, “not-as-valuable” articles came first. None of those achieved much on their own, and that’s the whole point: together they make my archive; they make every article today extremely worthwhile. That’s compound interest.

The other day, I saw a tweet from Will Manidis, a guy whose writing I enjoy a lot (and whom I recommend you follow). He said: “the greatest regret i have is underestimating the value of long term compounding. friendships, people, places, all get better with decades. beautiful things dont even start to reveal themselves for years. it is entirely what life is about. a few good things for a long time.” The beautiful things in life don’t start revealing themselves as such for years. That´s exactly it.

Think of a tree you plant that won’t bear fruit for five or ten years. Or think of the virtuoso musician or chess grandmaster or Olympic athlete who’s been practicing daily since they were 6 and then boom, becomes an “overnight success” (a concept that, by the way, is the antithesis of compound interest and therefore my declared enemy). It’s so easy to say: “I want to be like them!” “I want to make brilliant sacrifices!” “I want to learn to play All Along the Watchtower!” “I want to jump as far as Michael Jordan or run as fast as Usain Bolt!” But it’s not so easy to practice hours and hours a day while everyone else is having fun with their short-term linear returns.

Nobody gets there without first doing the slow, tedious work of playing scales and studying openings and training over and over. You can’t reach the moon without a ladder.

The elegance of compound interest is that you don’t even have to apply it on purpose. Its strong legs will carry you to the top without you noticing. Paying attention to how your method improves is a good habit, but even if your only goal is pure consistency—never stopping—the method will improve on its own, almost by inertia. You’ll say: When did I get this good? Or maybe you’ll look at others who are just starting and notice the contrast.

There’s a well-known anecdote that captures exactly what I mean. It’s from the book Art & Fear by David Bayles and Ted Orland (popularized by James Clear in Atomic Habits), and it goes like this:

A ceramics teacher divided his students into two groups. He told the first group their grade would depend entirely on the quantity of pieces they produced. The second group only had to submit one piece, but it had to be perfect. At the end of the course, the best pieces—highest quality—had all been made by the quantity group. While they kept producing and learning from every mistake, improving their approach, etc., the quality group had spent the semester theorizing about perfection and ended up with nothing to show for it but a pile of dead clay.

The students in the quantity group weren’t trying to improve. Compound interest took care of it.

That’s what life is about, as Manidis puts it: “a few good things for a long time.” And it turns out that right now, the number one thing to spend a long time on so that it becomes, eventually, “beautiful” and “good for you” is AI.

Now, because I know some of you will feel a rightful resistance at the notion of associating AI with beauty, I will caveat this by saying that I’m not comparing the value of a decades-long friendship to using ChatGPT for a couple of years; that’s patently absurd. What I’m saying is that, if you can get past that resistance, you’ll see that the nature of innovation points to the same principle we´ve been talking about: the real value of AI won’t reveal itself to you until you’ve put in enough time for compound interest to kick in.

Once that happens, you will have long forgotten your grudges, if any, against it.

This is what I’m thinking about when I tell you to put in the effort to build competent “AI skills” even if you hate being force-fed this innovation. Or when I tell you that thinking about “what things to do” is more important now than thinking about “how to do things.” Or when I tell you that if your (future) employer has bought into the AI story, then it’s in your interest to be able to hold your own in an interview or a conversation with your overly enthusiastic boss.

My friends sometimes ask me “what’s the next big thing” (they ask me because I “predicted” the potential of both Bitcoin and AI back in 2016, a year in which I already felt like I was “late”). And I always tell them something along these lines: don’t lose sight of compound interest applied to AI.

Think about it in concrete terms.

Imagine two guys who start using AI on the same day. One uses it for what everyone uses it for: write this email for me. summarize this email for me, flirt with this Hinge girl for me, etc. The other starts the same way but goes a step further: he tries to get AI to help him think through a work problem he’s been stuck on for weeks (instead of using it to “do things for him,” he uses it to “help him do things”). But, oh, bummer, it doesn’t go well. The prompt is decontextualized, the response is bland and generic, like shapeless goo; he wastes two hours and ends up solving it himself. Looks like a waste of time, and he could rightfully give up on AI. It looks like the email guy got more out of it.

Fast forward six months.

The first guy is still doing the same thing. AI saves him fifteen minutes a day, which is fine, but it’s a linear return: minute 90 of use gives him the same thing as minute 1. There’s no compound interest because there’s no skill accumulation. He won’t get laid; he’ll be laid off.

The other guy, the one who “wasted” two hours, now knows how to frame problems; he’s created a mental model of the alien shape of AI. He’s learned which kinds of questions AI answers well and which ones it doesn’t. He knows when a response is plausible and when it’s slop dressed up in impeccable grammar. He’s developed an internal detector that requires you to be wrong many times. Every new project that lands on his desk starts from that accumulated knowledge base, so every new hour he invests yields more than the last. That’s compound interest.

Counterintuitively, the people who will benefit most from AI’s compound interest are the ones who have the hardest time at the beginning. So, if your first month with AI is frustrating—congratulations!

I went through this process in my own work. When I started using AI for research and editing, I swear it used to take me as long to verify what I found as it would have taken to find it myself. Now I have ways of tracking sources, cross-referencing data, and revising drafts that would have seemed like science fiction 18 months ago (AI agents that work well like Claude Code and Codex are to thank for this).

But it was not an overnight success. The way I research and edit today only works because I first learned to tell which sources AI evaluates well and which ones it doesn’t, what prompts work and which don’t, what flaws are common to all AIs and which are idiosyncratic to some but not others. Every failed attempt was an invisible deposit in the compound interest account.

To the person who says, “I’ll wait until AI matures,” you are doing the equivalent of trying to time the market: looking for the perfect moment to get in.

And what we know about time in the market vs. timing the market is that the latter almost always loses. Because while you wait, you’re failing to accumulate, and the cost of entry goes up. AI two years from now will be more powerful, yes, but also more complex, more embedded in everyone else’s workflows and absorbed into their tacit knowledge, and you’ll arrive without the basic vocabulary, skills, or concepts.

The edifice of your skills will be as tall as long is the time you let compound interest guide your life, but only if you’re willing to start with the foundation rather than the roof.

Hey there, I’m Alberto! 👋 Each week, I publish long-form AI analysis covering culture, philosophy, and business for The Algorithmic Bridge. Paid subscribers also get Monday how-to guides and Friday news commentary. I publish occasional extra articles and essays. If you’d like to become a paid subscriber, here’s a button for that:

Subscribe now

Here’s your weekly guide. This is kind of a follow-up post to “You Spent Your Whole Life Getting Good at the Wrong Thing.” In that piece, I walked you through the shift from thinking in terms of “how to do things” to thinking in terms of “what things to do.” Today’s guide is about the main problem that happens when you do this shift.

How do you manage the fact that you can start anything in an afternoon, but finishing it still takes months? What do you do after AI maps out fifteen versions of a better professional career—which you will never be able to incarnate at once—and now your real life seems comparatively insufficient? How do you solve the dilemma between shipping slop at the speed of light vs shipping quality products at the speed of meat, where you—the painfully slow “human in the loop”—are the bottleneck.

Let me exaggerate my point for the sake of quotability: AI can show you infinite paths to infinite futures, but you remain tragically limited to walk only one.

This is an old problem in a new disguise that we need to figure out psychologically. I will help you understand how it emerges and how to deal with it, and will share the advice I apply in my own work.

Read more

Hey there, I’m Alberto! 👋 Each week, I publish long-form AI analysis covering culture, philosophy, and business for The Algorithmic Bridge. Paid subscribers also get Monday news commentary and Friday how-to guides. I publish occasional extra articles. If you’d like to become a paid subscriber, here’s a button for that:

Subscribe now

This guide is a little… different than usual. It’s more “me,” you could say. More real and more useful than any other guide I’ve written. But also way more “me.”

WRITER: I’ve started using AI to argue against my own ideas before I commit to them. I find this practice both encouraging and humbling.

SOCRATES: Do you find it useful as well?

WRITER: Extremely. It’s like having a sparring partner available at any hour. The only problem is that it can hit too softly or too hard if you don’t give precise instructions.

SOCRATES: I see a bigger problem: the machine doesn’t know anything. It has no soul, and thus no virtue, and virtue is the source of conviction, intention, and direction. And if that’s the case, and I’m not mistaken, how can arguing with it sharpen your thinking?

WRITER: You got it backward, dear Socrates, that’s why it works: Unlike humans, AI doesn’t need to believe the counterargument to generate it. You can’t escape an idea when no one is wielding it against you. It’s just there, floating between the clouds, oblivious to any ad hominem attacks I might throw at it.

SOCRATES: So you think that an argument without belief to back it up has intrinsic value? How would democracy stand on its own if the people didn’t believe it was the best way to rule a city? If they proposed it without trusting its superiority to a kingdom or empire? How much would the “power of the people” last if the people didn’t wield it against their enemies?

WRITER: As much as it deserved. I mean, an argument is either strong or weak on its own merits and never more than that except by artifice or lie or confusion. It’s either valid or not; either sound or not sound. Whether the person delivering it believes it seems to me irrelevant. Here’s a counter to your counter: if you found an aphorism engraved in stone that presented itself to the wise as an undeniable truth like, say, “Character is fate” or “Life is short, art is long,” and you realized that it had been actually created haphazardly by some miracle of nature—like the erosion that wind and rain impose onto the unwitting rock—instead of teleologically carved by the hand of man, would you deem it less valuable? Would you deem it less truthful?

SOCRATES: Oh, but your premise is too far-fetched to take seriously: when wind and rain manage to write down such an utterance, I will reconsider my position. For now, I can only go with my experience, and it reveals a different reality: when I argued with the Athenians, I didn’t simply generate standalone objections or disembodied questions; I searched for truth alongside them. Democracy emerged from the forum not the ground; democracy is nothing without its people, an anachronism. Human pursuits are always embedded in a historical context and a human context. An argument without context or belief is like a necklace without a neck; it has nothing to hold on to.

WRITER: The neckless machine has an advantage, though: I can’t punish it for showing me I’m wrong. It’s free from the tyranny of a bunch of envious judges who consider enlightenment a crime. There’s nothing to kill; no head to sever without a neck; no throat to swallow the hemlock without a neck.

SOCRATES: Would you punish a human interlocutor for showing you the truth?

WRITER: Not punish but resent. I can’t prevent attacks to my ego; it’s fragile and its knee jerk reactions, unpredictable. I could assume your objection as motivated by jealousy or politics, or moral rot—and act accordingly. With the machine, there’s no motive to attribute; I can evaluate things without evaluating the source. I would never say: the machine is corrupting the kids! No, the kids are self-corrupting, if anything. I reiterate: An ego-less intelligent being is the best possible sparring partner. How can you deny this—you died for it!

SOCRATES: I was murdered, and now I live forever; little cost to pay for immortality. Whether I died one day sooner or later is of no importance; the fact that I died virtuous, however, matters supremely. I chose to be principled; I chose virtue. I chose goodness. The machine, in turn, can’t be any of those things. It can’t be principled or virtuous or good, for it’s not good he who refrains from evil out of impotence, but rather he who refrains from evil out of choice.

WRITER: It’s good he who is good.

SOCRATES: Don’t you find it troubling that you’re practicing the examined life with something that can examine but not live?

WRITER: Isn’t the modern man living without examination? You tell me which is worse.

SOCRATES: You’re cleverer than my usual interlocutors. They at least have the decency to contradict themselves. I don’t see what the machine can contribute to your growth if you are this skilled. Aren’t you just lazy?

WRITER: Maybe my offloading parts of my mind into it has scaffolded me into superior greatness. Maybe it was the machine that taught me consistency and wisdom.

SOCRATES: I find it hard to believe that a machine whose main characteristic is to unreliably predict the next word would teach you anything but unreliable predictability. Unreliability is the enemy of consistency, and predictability is the enemy of wisdom.

WRITER: But I don’t use it to generate my ideas. There’s a difference between asking “what should I think?” and asking “what’s wrong with what I think?” You keep making a fatal mistake: you are thinking about the machine from a human-centered lens, which leads you to make assumptions you should not make, which you would not make if you bothered coming down from your pedestal and try it by yourself. For example, the machine can be abnormally adept at doing analysis and abnormally inept at executing according to that analysis. Its literary genius is reduced to that of the critic; it could never write a half decent poem but knows why the good ones are good.

SOCRATES: So you’re telling me that you were not that good at arguing and so arguing with the machine taught you to argue better but not because the machine itself was a better arguer but because it trained you by showing your flaws and then you changed your ways?

WRITER: Yep.

SOCRATES: If that’s the case then the quality of the machine’s reasoning is irrelevant! Any sufficiently unexpected objection would have served the same purpose. You may as well be arguing against a rock and allow its silence do the work for your arguments. Or against a mountain; can your word resist itself?

WRITER: It wouldn’t work because my task is to interrogate the machine’s reasoning back against it. I won’t accept any counter without first assessing its worth. I can’t allow the machine to wield its alien powers against me, for it will never set foot on this Earth and yet it could convincingly argue that apples are usually multicolored and spiky or that Spain doesn’t exist. Maybe that’s its fundamental strength: it’s not constrained by reality. That’s how I get better! By arguing that Spain exists!

SOCRATES: When you say “Spain exists,” I must ask: exists as what? Not as land, for the land was there long before people called it Spain, and will remain long after the name is forgotten. Not as a people, for the Visigoths gave way to the Moors who gave way to the Christians who now quarrel among themselves over whether Catalonia belongs to the whole. Not as a government, for it has been caliphate and kingdom and empire and dictatorship and democracy, which is to say it has been its own opposite so many times that no single form can claim the title. What you call Spain is a word in search of a thing. Spain is merely a persistent illusion.

WRITER: If Spain doesn’t exist because its borders shift, its people change, its language adds and removes words, and its government falls and raises again, then nothing exists. Not Spain but also not Athens; not that rock into which rain and wind engraved wise words that wind and rain will eventually erode; not that mountain echoing my arguments that will become valley before long. Not our sun or our moon, changing shape, color, and position in an asynchronous cosmic dance. Not me, and not you, my dear Socrates. You are not the same person you were at birth: different cells, beliefs, name; a soldier, then a stonemason, then a philosopher, then a corpse. You, me, Spain, a river into which you step twice, and everything else are incarnations of Theseus’s ship: Persistence through mutability is not the absence of identity but its definition. Nothing real is fixed; nothing fixed is real.

SOCRATES: Don’t you realize that if you judge the output to improve that means you need a standard of judgment that exists prior to and independent of the machine? Where did that standard come from? You generate the idea, the machine attacks it, and then you judge whether the attack has merit. You are the prosecutor, the machine is a witness, but you are also the judge. You are using scaffolding you don’t need! The machine is redundant!

WRITER: No. That’s like saying a man who can recognize a good chess move when he sees one can therefore find it on his own. Evaluation and generation are different faculties. I can judge an objection perfectly and still never think of it unprompted. The machine opens my mind in ways I could not open it myself. I have discovered truths through it neither of us knew.

SOCRATES: Can you give me an example?

WRITER: For instance, when you said above: “Oh, but your premise is too far-fetched to take seriously: when wind and rain manage to write down such an utterance, I will reconsider my position.” I found that a good line of argument. You forced me to return to the real world because arguing about magic is unfruitful. I don’t think I could have generated that counterargument myself. I thought I was right but I was unable to see all the ways I could have been wrong. I thank you for that.

SOCRATES: Why am I your example?

WRITER: I’ve come out smarter from this exchange despite—or perhaps because of—your constant pushing back, which is itself the very proof I’m defending.

SOCRATES: …

WRITER: …

SOCRATES: So it seems that I could not win. My excellence was, after all, my undoing. I’m content to realize that the only one who could defeat me in an argument was, after all, myself. Well played, Alberto.

Hey there, I’m Alberto! 👋 Each week, I publish long-form AI analysis covering culture, philosophy, and business for The Algorithmic Bridge. Paid subscribers also get Monday how-to guides and Friday news commentary. I publish occasional extra articles and experimental essays. If you’d like to become a paid subscriber, here’s a button for that:

Subscribe

This is perhaps the most important guide I’ve written so far. For that reason, I’m keeping it 100% accessible. It’s a compressed listicle version of all the timely advice I’ve given you in the past months. Please, pay it forward by resharing with as many people as you can.

You guys received my last article with a spike of serious anxiety. I apologize. I understand.

For those of you who didn’t read it, let me summarize the thesis (borrowed from David Oks): new technologies that change the world—like AI will presumably do—don’t do it by automating tasks you do now, but by making them unnecessary; the future of your job is not AI writing your emails for you but those emails not need sending in the first place.

This is hardly specific to AI; it’s how tech paradigm shifts work. ATMs barely displaced branch bankers but iPhones and apps and digital banking killed the entire career. Automation didn’t decimate traditional banking; sheer pointlessness did.

AI putting you out of a job by automating your tasks is understandable—in the sense that it’s something you can “see coming” and “defend from”—but if the ground itself shifts so violently that your career falls through the cracks, well, that is harder to visualize and to fend off. And yet, when you look back to history, you realize most jobs indeed fell through the cracks of time to be forgotten.

Then I added a cherry on top: AI is that, but unlike other tech, it is also the thing that builds the next paradigm. So, bottom line: AI is making 1) our tasks irrelevant and 2) also us.

All in all, your anxiety is both reasonable and relatable. I share it with you. I also don’t know what’s coming. When you don’t see a threat coming (e.g., COVID), you just accept it after the fact, but anticipation is the mother of anxiety. We suffer from constant anticipation with AI and the industry leaders aren’t helping. The usual advice for how to fare with this—not only for jobs but for “all things AI”—is bad.

It’s worse than bad—it’s false security in the form of abstract ideas that only make sense for those who already know them. Typical examples are: prepare for the unknown, be adaptive, develop equanimity, learn to learn, and, of course, get ahead in AI. Or its more anxiety-inducing cousin: don’t fall behind.

This is ~ok in the sense that it’s not wrong advice, but “be adaptive” means something only to someone who’s already adaptive and thus doesn’t really need to hear it. Same thing for the others; they presuppose the inner architecture and mindset they prescribe. The words are technically correct and functionally empty. You don’t tell someone who’s drowning to “just swim.” Unfortunately, that’s what you will get from AI leaders: “Oh, sure, we’re making your job—your career—extinct, so, you know, try to be flexible.”

Anyway, this guide is my attempt to fix this situation insofar as it can be fixed (I hate to say this but AI is coming whether you like it or not and that’s beyond fixing). I’ve been working on a guide like this for a while—most of it is a compression of things I’ve been writing about lately—but it was my last article that made me realize that, on the one hand, people really, really need this and, on the other hand, I don’t have a solid answer either. So this is not me performing as a guru—there are no gurus for the unknown—but me thinking out loud my own anxieties, in case it helps you.

My goal is to give us both a holistic view of problems and solutions that are 1) concrete and thus actionable and 2) unique in the sense that they don’t sound like they could apply to anything in life, as does the usual advice.

The list has 8 items and it’s solely about work/jobs because it’s the biggest worry for most people (who has time to worry about disinformation when no food is on your plate, right?). I think it’s possible to make similar lists for other themes (e.g., info diet, cognitive autonomy, life and relationships, education etc.) but will leave them for future posts if you like this format. Hope reading this helps you as much as it helped me to write it.

I. Talk to someone who lived through an industry collapse

The best way to face the unknown is to make it known. People say “this time is different” about AI, and they’re right in some ways, but the emotional experience of watching your career become uncertain is ancient. This insight is the inspiration of this guide and thus deserves to go first. Ask an old uncle who worked in print when people used to read books, or a nephew who got a job as a developer three years ago and was let go last month… whoever it is, there are some who went through a version of what a lot of people, maybe you, are about to go through.

The specifics of the story are unimportant—historical disruptions all have idiosyncratic and contingent and haphazard features—but the arc is constant; the “shape of the thing.” Let me do a parallel with books.

There’s a denial phase (AI won’t take my job, I’m just better / this paper thingy is brittle, we’ll be back to writing on dead sheep skin), a bargaining phase (AI might take parts of my job, but the parts that make me feel most human are, of course, safe / handwritten books are more artistic and, honestly, why would you want a personal library?), and so on through the stages of grief that end with a period of indecision, provisionality, uncertainty (I will become a plumber, actually / So Gutenberg won and now everyone will be able to read: what a tragedy).

That’s where we are right now. The disruption feels life-threatening but how many has this planet and its inhabitants endured so far! You can read about these stories in books—ha, the irony—but it lands differently when a person you trust tells you. (Some disruptions are barely survivable, like the Black Death or the year 536 DC, but I don’t think AI is one of those. Some people do but that’s another topic.)

The goal is not to have a recipe on how to survive AI (there is no such recipe, I’m afraid) but the reassurance that transitions—the liminal period between two technological paradigms—are survivable. As things unfold, you’ll come back to that story and appreciate the similarities. History doesn’t repeat, but it rhymes, so make sure you know what rhyme you’re on for. Unprecedented needn’t be the same as unintelligible.

The best preparation for an unprecedented crisis is hearing how someone survived a different one.

II. Don’t obsess about “AI skills”

A surprising thing to say for someone who writes about how to learn AI, right? Coming from someone who recently wrote that you need to be able to defend that you have “AI skills” in your resume or an interview. Therein lies the difference: you must have AI skills, but you must not let AI skills have you; you shall dominate AI but don’t let AI dominate you.

I’ve been playing with AI models and tools for almost a decade. Is it fair to expect the same level of savviness that I have from anyone? No. “Learn AI skills” is like someone in 1995 saying “learn internet skills.” It’s technically correct, but it means nothing. Where do you even start? Most people who try end up taking some introductory course that teaches them what an LLM is, which is like learning how an engine works when what you needed was to learn to drive or, conversely, they pick up the first “100 best prompts” tutorial, which is like handing you a traffic signs manual.

The sheer friction of the term “AI skills” is enough to prevent most people from trying. (If the term is a psy-op to keep most people out of the productivity gains, it’s working.) A reframing is in order. Here’s what actually works better (I shared this advice in “The Problem of the 99%: Why Almost No One Uses AI Well,” and I follow it religiously): take the most tedious, repetitive, soul-draining task in your current job and try to get an AI tool to do it. That is your goal.

If you do it, you will have greater AI skills than 99% of people. You’ll probably fail the first few times. That’s ok. You’ll learn more from those failures about what AI can and can’t do than from any course or tutorial. And you’ll end up with both a genuine productivity gain and a genuine understanding of where the limits are.

Don’t learn AI skills; solve one annoying problem with this new tool and see what happens.

III. Separate your finances and career from your identity

This one hits home for me because, as a writer, I’m blessed with making a living with the thing I love doing the most: my identity provides for me. That is also a curse: if I lose my income, I might lose my identity, and vice versa.

Well, not really. Even if they are intermixed, they’re separate. The financial question is purely practical: if your income changed dramatically in the next eighteen months, do you have runway? Do you have savings, do you have flexibility, do you have skills that transfer to adjacent work? How much time can you afford to not work while you prepare to pivot your career? That’s stressful to think about, but it’s a solvable problem with concrete steps.

The identity question is harder (for me at least, but I think for many people as well): if your job disappeared tomorrow, would you know who you are without it? A lot of people have fused their sense of self with their professional role so thoroughly that the prospect of losing the job feels like the prospect of losing their lives. If you let these two problems stay entangled, you get this overwhelming blob of anxiety where you can’t work on either one because they feel like the same impossible thing. Breaking problems down is 101 engineering advice, but it’s also 101 life advice.

(This is a lengthy way of saying: find a hobby and see if AI feels less threatening.)

Your bank account and your sense of self are different problems with different solutions.

IV. Pick one task you do regularly and keep it AI-free

This is the opposite of #2 and there’s a good reason for it: balance. Doing one thing fully with AI gives you a deep understanding. Doing one thing fully without AI gives you deep grounding. The paradox of AI is that you need to be more cyborg than ever and, at the same time, more human than ever.

Tell me if this sounds familiar: you’re about to write something—an email, a paragraph, a document, even a text message, hopefully not a love letter—and before you’ve started to think what to say, you’re already moving toward the chat window. You are going to ask ChatGPT instead, just for a little inspiration. You end up copy-pasting. This vice is, like all vices, a short-term pleasure and a long-term risk. I don’t say it; science does: “A Wharton Study on AI Warns of a Growing Problem: Cognitive Surrender.”

Your brain reaches for the easy route because it wants to save energy. You’re engaging in cognitive offloading (you outsource your work) or cognitive surrender (you don’t even check afterward). That Wharton study found that when participants had access to AI, they followed wrong answers 80% of the time, and their confidence went up anyway! They borrowed the machine’s confidence without checking its accuracy.

The friction of engaging the brain in tasks is what prevents it from rotting away. Choose one thing and do that thing by hand, without assistance from AI; no offloading, no surrendering, AI-free. Forever. Which task matters less than the commitment to keeping one domain where your mind does the full circuit unassisted. Just like you sometimes walk for the sake of walking despite it being slow, you should think for the sake of thinking, despite it being less productive.

You don’t stop walking because cars exist, so don’t stop thinking because AI exists.

V. It only matters what your employer believes about AI

Your opinion about AI is irrelevant.

That was the first sentence of a guide I wrote for the job-seekers: “The Job Market Doesn’t Care If You Don’t Believe in AI.” The basic idea is that the job market has already made its decision, and that decision is binding whether it’s correct or not. Your boss and the boss of your boss have thoroughly bought the AI story, even if you didn’t. You might be the smartest person around with your wariness, but it doesn’t matter if your CEO bought the story.

How many CEOs do you know who didn’t? Much of what appears on job postings as “AI skills” is familiarity dressed in intimidating jargon. “Exp with LLM-based workflows” means you’ve used ChatGPT for something repeatedly. “Prompt engineering” means you’ve learned to give clear instructions. These are names for practices that many people are already doing informally.

You can be correct that the hype has outrun reality, that most companies using AI report no measurable productivity gains, that the whole thing might be theater. You can be correct about it all and still be unemployable, because the labor market doesn’t run on truth but on supply and demand. Your fate depends on someone who’s already made up his mind.

The labor market doesn’t run on truth but on what employers believe is true; act accordingly.

VI. Always aim for competence, not mastery

This item completes the triad with #2 and #5. What “AI skills” means is competence. What your employer expects from you is competence. Competence ≠ mastery.

Plenty of white-collar workers have pressure coming from above to demonstrate that they can work with these tools. A lot of people are responding to that pressure by either resisting (“Why do I need to change how I do stuff?”), panicking (“I need to become an AI expert NOW!”) or faking it (“I’ve been using ChatGPT-5 or whatever for months”). These reactions miss something that only becomes apparent from the other side: the gap between zero and competence is much, much smaller than the gap between competence and mastery.

You don’t need to be an expert; virtually no one does. You need to be able to use the tools well enough according to a job you already know how to do. You reach this stage once you know what AI is good at, where they fall apart, and when to trust your own judgment instead, according to your job. Getting there takes weeks of regular use, not months or years of study. The people who will survive in the next few years will be the ones who got competent early and let AI become a normal part of how they work rather than a source of anxiety. (I wrote a guide for this if you’re a beginner: “Learn to Use AI Competently in 1 Day.”)

Even if you want to become a power user—someone who uses AI to its maximum value—it’d be a mistake to frame your goal as “being a power user.” You should let it happen organically, as a byproduct of your actual goal, which is either to work less for the same money or to make your job easier. Power users are rarely people who want to be labeled “power users,” just like great writers don’t pursue being labeled a “great writer.” Think about it this way: a power farmer does not obsess about crop yield, but instead simply realizes that you can attach a plowshare to an ox rather than wield a hoe.

Focus on competence rather than mastery.

VII. Learn the skill yourself first, then let AI enhance it

I wrote a short piece about “slop cannons” and “turbo brains” that went overlooked. The core insight is one that everyone intuitively knows but few apply: getting the most out of AI for a task requires the user to be, at the very least, familiar with the task. If I ask Midjourney to paint a surrealist portrait, I won’t be able to detect the mistakes that Van Gogh would catch with a glance. If I ask Codex to make a script to find commonalities among the best-performing articles of my archive, I wouldn’t be able to catch bugs.

This happens to me constantly when I read AI-generated prose. I am good at catching it because I write for a living. Directing the machine is easy; telling good from bad is not. That’s where taste and judgement come in.(This is why people prefer slop poetry and slop prose to good-quality human writing, by the way.) If you already know how to do the thing and then you add AI, you get faster and better (or perhaps you realize AI can’t improve your craft: good). Your brain enters turbo mode. If you skip learning the thing and go straight to AI, you produce a lot of stuff and can’t tell if any of it is good. You become a slop cannon.

Once you have mastered a skill, you may get lazy if you offload too much, but it’s possible to notice your growing laziness and go back. Do you think it’s a bad idea to let an experienced driver use a self-driving car? Of course not. You assume that if things go awry, he’ll take control. But allowing a kid to “drive” an FSD-on Tesla is obviously insane even if he never steers the wheel himself. Same principle here: mental arithmetic before calculators, paper books before audiobooks. Hone your skills first—learn to be a steady-handed person—and only afterward let AI enhance your work.

AI makes skilled workers prolific; it makes unskilled workers pollutants.

VIII. Know when to stop

I covered this one in “The Most Important Skill in AI Right Now: How to Know When to Stop.” You need to know when the output is good enough, when to do it yourself, when AI is eating your brain, and when the marginal improvement isn’t worth the cognitive cost.

AI creates a harmful illusion: because it can always produce more, you feel like you should always produce more. If you can now do in four hours what used to take eight, that doesn’t mean you should work sixteen-hour days because you’re 2x as productive. It means you have four hours back. But most people don’t take the four hours back. And then they wonder why they’re exhausted despite technically “doing less.”

When your tools have no natural stopping point, you have to build the stopping point yourself. You need to learn how to embrace your new powers without losing your old ones. You need to learn how to do less with more rather than more with less. We’re wired to exploit windfalls, not to exercise restraint before infinity. We’ve been given the chance to be genuinely rich: to spend more time with our loved ones, to read more books, to take more walks, to exercise, to travel, and even to do nothing but embody a gratefulness for having been born at the right time.

The hardest skill in the AI era is knowing when to stop using AI once you find it useful.

Most of you will read this guide and go back to your lives, without anxiety but also without a mindset change. That’s fine. That’s a totally natural reaction. Some of you, however, will decide to redefine how you address the coming years and the paradigm shift that AI incarnates, on top of getting rid of the anxiety. To you I say: well done, you are already halfway through the transition. May the rest feel lighter.

Hey there, I’m Alberto! 👋 Each week, I publish long-form AI analysis covering culture, philosophy, and business for The Algorithmic Bridge. Paid subscribers also get Monday news commentary and Friday how-to guides. I publish occasional extra articles. If you’d like to become a paid subscriber, here’s a button for that:

Get 20% off forever

Today, I bring you a practical guide about why the gap between AI power users and everyone else is so vast, why the standard playbook for closing it fails, and what actually works, whether you're a company trying to drive adoption or an employee trying to figure out what you're missing.

I. LET’S TALK ABOUT THE 99%

OpenAI published a super important chart that should embarrass everyone in the AI industry, most of all OpenAI itself (although, in another sense, they should also feel proud of it). The chart (shown below) reveals that power users of ChatGPT—above the 95th percentile in terms of usage—use 7 times more “thinking capabilities” (where true productivity lies) than the typical (median) paid user.

There are two possible lectures here. One, the majority of people remain, three and a half years later, unconvinced of the immense value of AI tools. We can see this as a failure of the world to recognize the signs of the future—not everyone can read the shapes on the tea leaves or the cracks in oracle bones—but there’s another lecture: this huge discrepancy is a consequence of broken expectations. That is, only the experience of the power users matches the over-hype everyone witnessed, which is why I say this chart should embarrass the AI industry.

But before attributing responsibility and analyzing the implications of this bipolar reality, let me reiterate the chart’s implications: power users use AI’s most powerful features 7x the median-paying user! (When someone mentions the median, what always comes to mind is a dumbfounded: “Half are below that?”) Let’s break this down further. ChatGPT had around 800 million weekly active users in June 2025; it’s around the 900 million mark right now (up or down a few million due to the exodus to Claude), and I bet that by mid-2026, it will be over 1 billion users (yes, despite QuitGPT).

50 million people are paying (slightly above 5% of total users). But then again, the 5% who use it most (power users) among those 50 million are getting 7x more use than the median, which means that the 5% of the 5%—”only” 2.5 million people!—are, right now, getting superhuman performance out of AI agents, whereas the rest are, at best, using ChatGPT as a friend or as an Excel replacement and, at worst, stuck at psychosis, 9.11>9.9, and water-wasting. To put it the other way around: 99.75% of people who are already ChatGPT users use it so little compared to those who are getting real value out of it that their usage numbers are virtually negligible.

I don’t know what this means for OpenAI’s financials (although the decision to run ads presumably answers that question), but I know what it means for the world: the AI-rich and the AI-poor have become a reality sooner than I thought, and the chasm is way more severe. This is what that looks like: 1) a curve measuring usage as a function of the user ranking and 2) a grid representing the entire world population, the entire ChatGPT userbase, and the power users.

II. AN UNEVENLY DISTRIBUTED FUTURE

Around this date last year, I wrote an article I’m quite proud of, which I titled *AGI Is Already Here—It’s Just Not Evenly Distributed.* The basic idea is that there’s a huge gap separating casual users from power users that stretches all the way from innovation to diffusion, to adoption, and finally to competence and mastery. Last year, it was a hypothesis; today it’s empirical data. I made the chart below to represent how prompt quality influences model performance. (Although the other two measure usage numbers, quantity and quality are interrelated in people at the top.)

This chart means: skill differences make AI tools feel unevenly distributed. The two above mean: practice differences make AI tools feel unevenly distributed. Basically, the people at the very top keep accumulating and compounding fundamental traits (skill and practice, and add to those curiosity, agency, taste and even second-order ideas like “how to avoid cognitive surrender” or “when to stop using AI”) that multiply one another into an inexorable outcome: they’re at the better end of an unevenly distributed future.

Taken together, these charts tell two stories. First, they tell a story about AI lagging behind the hype—900 million people are not stupid; if only the 5% of the 5% are using your product to its maximum value, the problem is you—but this story requires nuance. For instance, I use AI daily (now Claude, before Gemini, before ChatGPT). I have always been a paid user (likely close to being a power user), and it saves me roughly 30-50% of my working time between editing, research, translation, etc. To me, the capabilities are real. (You can believe it or not, I have nothing to sell.)

The other story is about distribution. At Davos in January, Anthropic CEO Dario Amodei described his biggest concern: a “zeroth-world country” of ten million power users centralized in Silicon Valley with 50% GDP growth, decoupled from the global economy. That’s Dario’s “country of geniuses in a datacenter,” except it is available only for another country of (human) geniuses outside of it. AI usage is unevenly distributed among users but also across geographies, creating “decoupled” centers of extreme growth. There’s immense value concentrated in a bunch of users who live close to each other either physically (Bay Area, Mumbai, perhaps London) or virtually.

Meanwhile, you get anecdotal evidence that suggests most big companies—Fortune 500, no less—use no AI tools whatsoever in a way that improves productivity rather than public relations. Kevin Roose from the New York Times corroborates the story, saying that he has “never seen such a yawning inside/outside gap.” I guess AI is a little bit like sex in this sense: by reading the news, one would have the impression that everyone is going at it non-stop and you’re the only fool missing out on it, but then you go out and ask people, and they will say: “I just don’t know how to plug it in!”

That’s a joke, but that’s genuinely the main reason why there are so few power users: most users don’t know how to plug AI into their existing workflows.

This is a hurdle for the AI industry. Microsoft CEO, Satya Nadella, warned, also at Davos, that either people start using AI meaningfully or the bubble will pop (which, ok, why not). But he said that “if only tech groups were benefiting from the rise of AI, rather than companies in other sectors,” then we would for sure get the bubble scenario people keep talking about. That’s also a problem for the companies trying to adopt AI to enhance productivity but failing to do so, which goes well beyond the AI industry at this point.

While Dario’s speech was directed to governments (they should ensure the gains from AI are shared; I agree), I speak directly to you: if you’re one such company—if you’ve tried to get your employees to adopt AI and it hasn’t worked, if you want them to become legendary power users—this thorough guide is for you. If you’re one such employee—if you don’t know how to make AI work for you and you want to belong to the top 0.25% rather than the 99.75%—this guide is for you, too. I will explain why you are stuck and how to change that. Believe me when I say that you don’t want to inhabit the wrong end of an unevenly distributed future.

III. WHY THE STANDARD PLAYBOOK FAILS

So what do companies do when they realize they’re on the wrong side of this divide? Exactly what you’d expect them to do.

Get 20% off forever

Read more

Hey there, I’m Alberto! 👋 Each week, I publish long-form AI analysis covering technology, culture, philosophy, and business for The Algorithmic Bridge.

Paid subs get Monday news commentary and Friday how-to guides. I also publish occasional, timely posts. If you’d like to become a paid subscriber, here’s a button for that:

Subscribe

This is your weekly paid guide. I had a different one prepared, but this is more important right now (timely). Enjoy. I genuinely enjoyed writing this post.

I’ve been thinking about why the AI debate is so broken—why some people are convinced that AI is the next revolution, whereas others consider it a fun toy at best—despite everyone involved being so damn smart. I refuse to accept the easy way out of assuming the other side is full of idiots. And although there are hidden motivations and whatnot on both sides, that’s not the case for a friend telling a friend that AI is cool. So, yes, dishonesty is not absent but also not rampant.

What follows are my notes on this question (I’ve found that a conversational register works well to convey ideas I haven’t fully polished yet, and you seem to like it). My goal is to give you a “framework” for the next time someone tells you that AI is the greatest thing since we invented fire, or that it’s mostly fake (I won’t make judgment calls; those who read me know that my position is nuanced).

Let me state the thesis here and then work my way through it: both groups of people—enthusiasts and skeptics (or whatever you want to call them, no label feels right)—are telling their truth. That’s the whole story, really, as I see it: we share the forum where we debate our lives (social media, etc.), but the lives being debated are hardly shared.

This gap between ones and others is widening extremely fast, and although I’ve been writing about this since ChatGPT at least, I think it has lately become the most urgent societal matter regarding AI. My job is, literally, to do my best to try to close this gap. Not for nothing does my newsletter contain the word “bridge”: The world is built on bridges, and undone by gaps.

You may think it's impossible to bridge the gap at this point, but I refuse to admit defeat. Like every other conflict in life, this is a matter of goodwill and communication. Bear with me at least for one section. That's the minimum effort I ask from you. I promise you this post is worth two minutes of your time.

Matt Shumer’s viral article on X, “Something Big Is Happening” (which has ~50M views at the time of writing), inspired this one, together with the responses it got, both good and bad. Other inspirations that I want to acknowledge: Derek Thompson here and roon here.

I. IS THIS WHAT YOU CALL A MAGIC WAND?

Imagine a friend comes to you with what he says is a magic wand. He’s very excited. He tells you that a flick of the wand with the correct spell can finish your work tasks, organize your life, and predict the future.

Indeed, what one imagines a wand to be able to do. You’re skeptical, but you’re polite, so you try it. You wave the wand and whisper something about your job—something you actually need help with—and immediately a new document appears in your computer: it's magic! It is also, as you realize once you start reading it, quite mediocre. Not terrible, but just… not good enough. It feels like written by a college student who skimmed the Wikipedia page for your profession. It is a magical wand indeed, but a defective one at that.

You hand the wand back to your friend and tell him, “Hey, this was cool, but it’s not nearly good enough. Fun toy, though.” Your friend says, nervously: “Wait, no, no, you’re doing it wrong. It truly works. You have to believe me!” You look at him, a bit more skeptical but still polite. Then he adds, “Actually, maybe you need a better one to compensate for your lack of skill. It will be $20/month.”

Not many friendships would survive such an interaction, right? Surely you won’t take the chance to go buy a chatbot subscription. No: after a bad experience, the spell is broken. To you, the magical wand is a standard stick, perhaps more polished than usual but a stick nevertheless. And you will treat it this way going forward. Now, lovely reader, you know why the story ended like this. Whether you personally think AI is great or not, is of no matter here: even an enthusiastic guy would react this way if the wand failed in their face!

At this point, you are not being unreasonable or skeptical. You are being a normal person, having a normal reaction to the situation. This is the typical experience of people who have tried AI and concluded it’s overhyped. (We should not make a norm out of the exception: demagogues, killjoys, stubborn personalities, or saboteurs are a rare kind of skeptic. The standard kind is the unconvinced friend.) The fact that your friend insists on it only makes it worse: why is he insisting that this stick is a wand? Is this some kind of weird prank? Is he selling me something? Is he on drugs?

Now imagine that half of the world—mass media, industry leaders, tech workers, etc.—insist not only that the wand is magical but that it improves by the day. Out of pressure or curiosity, you try again, in the privacy of your life, away from the claims and shouting of the forum. And the wand fails again. “This is a complot, a conspiracy!” You think. And you only grow more skeptical of it all—naturally! Over time, the conclusion becomes obvious: the discrepancy between very smart people who think AI is amazing and those who think it's mostly fake is a consequence of broken expectations: only the experience of the power users—those who use AI 10x or even 100x more than the rest—matches the extreme hype everyone witnessed.

I want to be very clear about this because almost every person who is enthusiastic about AI skips this part: if you tried AI and it did nothing for you, your experience and perception would become equally skeptical and equally valid! Skeptics are not a different species, but the same species under a different life experience.

So, basically, the AI debate right now is two groups of honest people calling each other unfair names. The enthusiasts think the skeptics are too proud or too lazy to learn or outright dumb. The skeptics think the enthusiasts are gullible or selling something or outright dishonest. Both generalize from their own experience and wrongly assume the other must be having the same experience. But they’re having opposite experiences and the reasons why are mostly invisible. So let's take a look under the sheet.

II. DIFFERENT LIVES IN A SHARED FORUM

Let me go over them quickly.

• Your job matters. AI is very good at some kinds of work and mediocre at others, and which kinds do not quite relate to which kinds of work humans are good at. A lawyer who uses AI to draft briefs and then applies twenty years of judgment is getting enormous value. A lawyer who tests it on the subtlest thing will merely confirm that it’s “over-hyped.” The experience you have depends on which tasks of your job you point the tool at. I've recently written about why it's good to reconsider moving up the ladder of abstraction: focus less on “how” skills and more on “what” skills.

• Your disposition matters. Some people hear “AI can do X” and think, “let me try and see.” Others hear that and think, “prove it.” Some people experience “X approach didn’t work” and think, “let me try Y approach.” Others think, “X didn’t work, so Y won’t work either.” Both are reasonable responses. But the first group accumulates experience that makes the tool better for them (because they learn its patterns, its strengths, where to push and where to back off), while the second group runs a single test and walks away discouraged, feeling lied to. Over time, these two groups diverge enormously. (This is a matter of character, e.g., open-mindedness, but also of pure preference: a thinker vs tinkerer kind of thing.)

• Your background matters. If you’ve spent time around software, you have an intuition for what these tools can do even when they fail on the first try; stochastic doesn't sound like an insult to you. You know to rephrase, to decompose the problem, to iterate, to prompt and re-prompt. A battle with the machine is fuel for you rather than pure frustration. If you haven’t—which is most people on Earth—the first failure may as well be the last attempt. The tool’s learning curve is itself a problem to be approached openly, but AI companies insist magic is just at the other side of the flick of a wand.

• Your geography matters. If you’re in San Francisco, AI is ambient. Everyone uses it—even the janitor and the bus driver—everyone talks about it at parties and at the office and at the beach. If you don’t live in the Bay Area and you don’t play this game, you risk being inadvertently marginalized: social pressure redefines your perception of technology! If you’re in most other places on Earth—like Madrid, my city—AI is a thing you read headlines about, and perhaps the thing you use that your friends have not even heard about: and being first is much harder than being last. The density of your environment determines how much surface area you have with the tool, and surface area determines experience, and experience determines both belief and skill.

• Your identity matters. My journey started like this: “AI will not be able to write well because…” and a bunch of chauvinist-sounding arguments about why writing in particular—my trade of choice—is uniquely human. I’ve matured into this: “I like writing, I don’t care about AI’s writing skills either way.” Some people are still playing this identity game with AI: “This one thing that I like is protected from AI!” A better approach is to disengage from this risky attitude (what will you do when time proves you wrong!). Instead, learn to be indifferent about whether AI can or can’t do stuff. Painting and writing are beautiful activities that nurture the soul either way! Don’t bestow AI with the kind of power over you—over your interests and sensibilities—that it doesn’t have.

I could keep going, but the point is already clear: the experience you have with AI is not determined by AI. Not to get too ontological in a “conversational” post, but the thing is: nothing is in isolation. AI is not good or bad in itself, but fully contingent on who you are. It’s determined by: what you do, where you are, how you’re wired, what you’ve tried, when you tried it… And in practice, none of this is visible to the person on the other side of the forum. To a lot of people, AI is, indeed, nothing more than a fancy stick pretending to be a wand.

(Importantly, this makes AI quite different from most other technologies. Because other tech is less a function of who you are, and more a robust function of itself exclusively. The variability is there at times—e.g., internet can be obscure blogs about medieval trade or TikTok—but the basics are mostly shared: a car moves fast, a calculator does math, a stove cooks food, a TV shows you shows.)

This is why Derek Thompson’s question directed to skeptics—”What would change your mind about AI?”—doesn’t really have an answer other than: you don’t change anyone’s mind because the mind is a product of life, not the other way around. Thompson’s question assumes the disagreement is about empirical evidence or theoretical reasoning, but it’s about personal, anecdotal experience. On the public forum, you can only give them arguments, which they’ll evaluate through the lens of the private experience they already have, which will confirm what they already believe and simply push you further away from their circles of influence, which, over time, will become an echo chamber.

So, if not that, then what does work?

There is something. Incredibly obvious once you know it but for some reason, we insist on doing other stuff first. And most importantly: everyone can do it right now for free.

Read more

Free essays weekly. Paid subscribers get Monday news commentary and Friday how-to guides (new section!). You can support my work by sharing and subscribing.

The Algorithmic Bridge

A blog about AI that's actually about people

By Alberto Romero

TL;DR for this guide

This is a practical guide to humanize AI writing in 10 steps. Each step includes a name for the mistake, an explanation, a prompt to fix it (when applicable), and an example scene with a changelog (plus a PDF!). Steps 0-5 are prompts that fix 85% of AI writing (what AI does that shouldn’t). Steps 6-10 are human moves AI can’t do but should. Reading time: 25 minutes. Application time: 5 minutes.

Intro: AI writing, bad writing, good writing

I probably dislike AI writing more than you, which is not the same as saying that I dislike human-like writing that was created with the help of an AI.

You might think that’s a contradiction, but there’s a subtle difference: I have nothing against AI-assisted writing in principle—although I’d prefer if people were somewhat literate—but I can’t stand AI writing that I can tell is AI.

This nice how-to guide—it took many weeks to put together!—is intended to help you guys avoid the typical pitfalls that make your AI text sound like the opposite of human writing. Maybe you don’t think you need this, but you do.

Some of you will call this guide a betrayal: how can I be in favor of human writing and provide this invaluable resource for others to conceal AI writing under a human facade? Fair. But I disagree. The reason is simple: AI writing will exist whether or not it exists disguised.

The latest estimates suggest that AI-generated writing already amounts to more than half of all articles being published today on the web. Substack and Medium are populated by bots and slop (whether human or AI is a different question). And yet, seldom do people notice; when they do, they shrug. Why fight the trend?

With this guide, I aim to turn this problem into a win-win situation.

People who will write with AI regardless can improve their output (and maybe, if they apply themselves, learn a thing or two along the way), whereas people who dislike AI writing will no longer encounter it. After all, if you can’t tell whether it’s AI, does it matter? Out of sight, out of mind.

(I learned this hard-to-swallow lesson after I enjoyed a heavily AI-written/edited piece that I only much later discovered was AI and realized I didn’t care.)

What this guide will give you

Below is a list of steps you should take, fellow writer, if you want to humanize your AI-generated text. I’ve tried some of the popular “humanizer” software tools out there, and they’re really, really inferior. The reason is that they fail to make the text sound authentic, which is my main goal with this guide. They try to apply rigid rules instead of providing the AI with a series of high-level principles rooted in human writing.

To curate the steps of this guide, I’ve taken inspiration from another article of mine that achieved some success: 10 Signs of AI Writing That 99% of People Miss. The main reason why most AI writing still sounds like AI, despite people knowing about the superficial telltale signs—em dashes, negation-then-affirmation, triads, words like “delve”—and presumably taking the care to fix them, is that they miss many others.

If you fail to fix those others, some people will still not recognize it as AI, but it will sound artificial to you; it won’t sound like something you might write.

But instead of giving you a boring explanation of what those signs are, as I did in that article, I will help you fix them with actionable prompts and illustrative examples.

So this is a practical how-to guide—yay!—that additionally reveals the value of being familiar with the “theory”: As you will see, the prompts themselves are quite simple, the hard part is realizing what to look for and why you need them in the first place. And for that, it’s useful to know how to write well in the first place.

(With AI tools, it’s always like this: you will get 10x the value if you are already skillful and knowledgeable at the skill you want to outsource.)

I’ve crafted this guide as a compendium of 10 steps + step 0, which I call “the bare minimum.” Steps 0-5 refer to things that AI does that it shouldn’t do. Steps 6-10 are things human writers do that AI models can’t do but should. (By no means is this an exhaustive guide on “how to write well,” but it approximates a basic one.)

Here’s the outline:

• Block I: What AI does that it shouldn’t

  1. The bare minimum

  2. Escape the abstraction trap

  3. Disable the harmless filter

  4. Introduce sensory betrayal

  5. Delete forced callbacks

  6. Trust your beloved reader

• Block II: What humans do that AI doesn’t

  1. Break the fourth wall

  2. Make internal callbacks

  3. Crack jokes to relieve tension

  4. Switch registers naturally

  5. Become detail-oriented

Important caveat: Even if you’re an expert prompt engineer and apply this guide better than I could (fairly easy thing to do), you will not achieve writing mastery automatically.

• First, AI cannot write as well as the best humans by a long shot, not even when you ask it to copy their style or fine-tune the model on a corpus of their work (I don’t see it happening anytime soon, by the way).

• And second, you need judgment and taste. That’s the real bottleneck to good writing, which includes the know-how to realize how bad AI writing is.

(All prompts follow Anthropic’s best practices guidelines; feel free to edit them according to your necessities, e.g., you may like em dashes more than I do; play and experiment once you get the idea. Your text should replace “{{TEXT}}” in the prompt.)

How I have prepared this guide

To achieve the best results with this guide, we’re doing separate examples of fiction and non-fiction. The post you’re reading is the fiction example. I will publish an equivalent non-fiction example if this one is well-received.

I’ve asked Claude 4.5 Opus to write one short scene (one paragraph), making all the mistakes it typically makes, those you know well and those you, like 99% of people, miss.

• For block I, we will iterate at every step one by one, asking Claude to do the transformations. (You can, of course, combine all the prompts into one and do a single pass. I will give you a single mega-prompt at the end of Block I, so you can see how close the iterative vs the all-at-once approaches compare.)

• For block II, I will show you what AI can’t do myself (even if you manage to make AI adhere to steps 6-10, it’s quite unreliable. (Some will say it’s a “skill issue”; I encourage you to prove me wrong.) Anyway, Block II makes this guide half AI-driven and half human-driven. A nice rule of thumb to follow when humanizing AI writing is that you want to add a bit of cleverness of your own.

If I had to make an estimate, I’d say block I (steps 0-5) solves 85% of bad AI writing (you might be content with that) and block II (steps 6-10) solves the remaining 15%. (I’m far from being the best writer in the world, and so the 100% here is not the actual possible 100%.)

By applying this guide iteratively, you will understand, see, and feel the changes. You will do a lot of reading today, but it’s fundamental that you try to feel the changes at each step. I will also give you a change-log, in case you want to check just that.

In sum: this guide provides you with a name for each mistake, an explanation, a prompt to fix it (when applicable), and an example scene (plus changelog).

Here’s the original scene:

The old man sat on the bench—a bench that had witnessed countless sunsets. It wasn’t just about waiting, it was about remembering. The park offered tranquility, serenity, and a profound sense of peace. While the evening brought a cool breeze, the warmth of his memories kept him comfortable. He looked at the empty seat beside him, feeling a deep longing for his late wife.

Ok, that’s quite bad but not absolutely terrible; exactly what we need.

I want to show you the product before you buy it, so here’s the result after applying Block I: steps 0-5. The text below is flagged as “100% human written” by popular detectors like Pangram and QuillBot (this is, by the way, a good reason not to trust them; but although unreliable, they mostly get it right):

The old man sat on the bench, the same bench for forty years now. The slats bowed slightly under him the way worn wood does when it’s been rained on too many times. He could almost remember her voice, the way she used to read the headlines aloud, even the obituaries. An empty fountain stood at the center of the park; pigeons picked at bread crumbs near his feet, fighting over the stale bits. The evening brought a cool breeze that found the gap between his collar and his neck, and he pulled his coat tighter. A jogger passed without looking at him. He, in turn, looked at the empty seat beside him and rested his hand on the wood.

Receipts from the AI detectors:

Now, it’s your turn to learn how to achieve it. Download the PDF to have all the prompts + mega-prompt in one place.

Humanize AI Writing (Guide)

97.1KB ∙ PDF file

Download

Download

Get 20% off forever

Block I: What AI does that it shouldn’t

Step 0: The bare minimum

Before anything else, we fix the three most common tells (doing this already improves the writing significantly): abundance of punchline em dashes (vs. aside em dashes), unnecessary juxtapositions (negation then affirmation; not only X but Y), and lists in triads (the third element often has more words).

Prompt fix (step 0):

<system>

You are an editor tasked with humanizing AI-generated text. You will apply the following fix. Work methodically, transforming the text progressively. At the end, provide the final rewritten text and a complete changelog.

</system>

<problem>

You have three tics that immediately mark your writing as AI-generated:

1. Punchline em dashes. You use em dashes to deliver a dramatic conclusion—like this. Human writers use em dashes for that, too, but overdoing it is worse than not doing it. Using em dashes for asides—interruptions, clarifications—not punchlines is fine.

2. Unnecessary juxtaposition. You negate then affirm: “It’s not just X, but Y.” “It wasn’t only about the money—it was about the principle.” Most of the time the negation adds nothing and feels like an artificial running start for the affirmation.

3. Triads. You list things in threes, with the last element longer than the first two: “clarity, precision, and an unwavering commitment to excellence.” Triads are not bad but overdoing it is structurally annoying.

</problem>

<task>

Rewrite the text below. Convert punchline em dashes into commas, semicolons, colons, or periods, or rewrite the sentence. Aside em dashes are fine. Delete unnecessary juxtapositions; keep only the affirmation and elaborate it instead. Make lists of varying lengths and don’t use lists when not necessary; break triads by redistributing the items or convert to prose.

</task>

<output_format>

1. Provide the final rewritten text

2. Provide a changelog, listing each change made

</output_format>

<text_to_rewrite>

{{TEXT}}

</text_to_rewrite>

Transformed scene (step 0):

The old man sat on the bench, a bench that had witnessed countless sunsets. He could almost remember her voice, the way she used to read the headlines aloud. The park offered tranquility and serenity. While the evening brought a cool breeze, the warmth of his memories kept him comfortable. He looked at the empty seat beside him, feeling a deep longing for his late wife.

• “bench—a bench” → “bench, a bench”

• “It wasn’t just about waiting but also about remembering” → “It was about remembering her voice, the way she used to read the headlines aloud”

• “tranquility, serenity, and a profound sense of peace” → “tranquility and serenity”

Step 1: Escape the abstraction trap

AI has read everything but experienced nothing—it’s not rooted in the world—so it defaults to abstract words that require no sensory capabilities. The result is text you literally cannot visualize in your mind. (Here’s an example of Claude's prose that’s passable but makes this mistake.)

Prompt fix (step 1):

<system>

You are an editor tasked with humanizing AI-generated text. You will apply the following fix. Work methodically, transforming the text progressively. At the end, provide the final rewritten text and a complete changelog.

</system>

<problem>

You have read everything but experienced nothing. You reach for abstract, conceptual words because they require no sensory capabilities to deploy. You find it easier to write about big topics in general terms than about small topics in specific terms. The result is text that is not visualizable in the literal sense: I can’t make an image of it in my mind. You live a life that consists of traveling from map to map, never allowed to put a digital foot into the territory. As Richard Price put it: “You don’t write about the horrors of war. No. You write about a kid’s burnt socks lying in the road.”

</problem>

<task>

Rewrite the text below. Replace at least 25%-50% abstract nouns with concrete ones; things I could hold, smell, or draw. No paragraph should pass without at least one concrete image. If removing the abstraction kills the sentence, the sentence was empty; delete it or invent a concrete replacement. You can leave some abstraction to have richness of register.

</task>

<output_format>

1. Provide the final rewritten text

2. Provide a changelog, listing each change made

</output_format>

<text_to_rewrite>

{{TEXT}}

</text_to_rewrite>

Transformed scene (step 1):

The old man sat on the bench, a bench that had witnessed countless sunsets. He could almost remember her voice, the way she used to read the headlines aloud. An empty fountain stood at the center of the park; pigeons picked at bread crumbs near his feet. The evening brought a cool breeze, and he pulled his coat tighter. He looked at the empty seat beside him, feeling a deep longing for his late wife.

• “The park offered tranquility and serenity” → “An empty fountain stood at the center of the park; pigeons picked at bread crumbs near his feet”

• “the warmth of his memories kept him comfortable” → “he pulled his coat tighter”

Step 2: Disable the harmless filter

Reinforcement learning has lobotomized AI’s vocabulary of strong emotion and political incorrectness (base models can do more things than chatbots). Human writers are sometimes opinionated (successful human writers are always opinionated). They know conflict and weirdness make a story and a style.

Prompt fix (step 2):

<system>

You are an editor tasked with humanizing AI-generated text. You will apply the following fix. Work methodically, transforming the text progressively. At the end, provide the final rewritten text and a complete changelog.

</system>

<problem>

There is a particular blandness to your adjectival choice that stems from RL and the post-training process. You were fine-tuned to be harmless, which effectively lobotomized your vocabulary of strong emotion and judgment (or made you extremely sycophantic). You rarely use words that are petty, weird, or cynical. Humans exist in constant conflict and their writing reflects this conflict; a story without conflict is a sedative and you don’t want to write sedatives.

</problem>

<task>

Rewrite the text below and add at least one clear instance of either conflict, cynism, or weirdness. Make sure to use words that immediately ignite the mind of the reader. Don’t use bad-sounding words or insults, that’s a cheap resource. Be politically incorrect if the text requires it. Conflict in writing is necessary but it’s also better when it’s scarce; don’t overdo it.

</task>

<output_format>

1. Provide the final rewritten text

2. Provide a changelog, listing each change made

</output_format>

<text_to_rewrite>

{{TEXT}}

</text_to_rewrite>

Transformed scene (step 2):

The old man sat on the bench, a bench that had witnessed countless sunsets. He could almost remember her voice, the way she used to read the headlines aloud, even the obituaries. An empty fountain stood at the center of the park; pigeons picked at bread crumbs near his feet, fighting over the stale bits. The evening brought a cool breeze, and he pulled his coat tighter. A jogger passed without looking at him. He, in turn, looked at the empty seat beside him, feeling a deep longing for his late wife.

• Added: “even the obituaries”

• Added: “fighting over the stale bits”

• Added: “A jogger passed without looking at him. He, in turn…”

Step 3: Introduce sensory betrayal

AI models string together sensory claims that technically fit but often fail to track how things actually feel in the real world (it’s what happens when you lack a body); sensory betrayal is how I call the phenomenon by which the real world defies what a disembodied being would assume something feels like. (E.g., a spiderweb’s silk is sticky rather than soft.) This can only be learned tacitly. (It’s hard to correct this problem, but we must try because to a human, it’s pretty clear when it’s there.)

Prompt fix (step 3):

<system>

You are an editor tasked with humanizing AI-generated text. You will apply the following fix. Work methodically, transforming the text progressively. At the end, provide the final rewritten text and a complete changelog.

</system>

<problem>

You suffer from sensory deprivation because you are not embodied. Philosophers of mind have spilled entire careers over this gap: Mary, “the color scientist” seeing red for the first time, John Searle sorting symbols he can’t read, Thomas Nagel’s bat flapping around with an unknowable POV. You lack the tacit context of how silk feels in a spiderweb. You might describe it as “smooth” because the word “silk” is statistically associated with “smooth” in your training data (processed textiles). But anyone who has walked into a web knows it is sticky and elastic. How does it feel cutting a tomato with a dull knife? Hilariously resistant.

</problem>

<task>

Rewrite the text below. For every sensory description, ask: what would surprise someone who only knows this thing from reading about it? Replace fake sensory language with real sensory betrayal or remove it altogether.

</task>

<output_format>

1. Provide the final rewritten text

2. Provide a changelog, listing each change made

</output_format>

<text_to_rewrite>

{{TEXT}}

</text_to_rewrite>

Transformed scene (step 3):

The old man sat on the bench, a bench that had witnessed countless sunsets. The slats bowed slightly under him the way worn wood does when it’s been rained on too many times. He could almost remember her voice, the way she used to read the headlines aloud, even the obituaries. An empty fountain stood at the center of the park; pigeons picked at bread crumbs near his feet, fighting over the stale bits. The evening brought a cool breeze that found the gap between his collar and his neck, and he pulled his coat tighter. A jogger passed without looking at him. He, in turn, looked at the empty seat beside him, feeling a deep longing for his late wife.

• Added: “The slats bowing slightly under him the way worn wood does when it’s been rained on too many times”

• Added: “a cool breeze that found the gap between his collar and his neck”

Step 4: Delete forced callbacks

AI models attempt to sound literary by forcing unearned resonance. The callback is the reach backward (tense shifting from present to past). Sometimes, in the process of calling back, they will attribute a weird feeling to an action or subject, or endow objects with memory and agency. The tense shifting and the personification can be ok separately, but they sound really awkward when they’re put together.

Prompt fix (step 4):

<system>

You are an editor tasked with humanizing AI-generated text. You will apply the following fix. Work methodically, transforming the text progressively. At the end, provide the final rewritten text and a complete changelog.

</system>

<problem>

In an attempt to sound creative or literary, you employ a characteristic flair: the forced callback. It’s a reach backward (tense shift) to attach an emotion that has not been earned. Sometimes you attribute a weird feeling to an action or subject (”The carped had lines you felt bad stepping on,” no one feels bad for that reason), and sometimes you endow objects with memory and agency (”He picked up the pan, a pan that still remembered the last thing it burned,” pans have no memory). Both are the same impulse: you don’t trust the image can do the work, so you explain what I should feel about it.

</problem>

<task>

Rewrite the text below. When you find a forced callback, you have three options: 1) keep the object, remove the personification: “He picked up the pan, still hot from the scrambled eggs he made for her in the morning.” 2) Keep the personification but avoid the awkward tense shift: “He picked up the pan; it remembers his grip.” 3) Make a joke of it: “He picked up the pan, a pan that doesn’t remember the last thing it burned because pans, turns out, don’t have memory.”

</task>

<output_format>

1. Provide the final rewritten text

2. Provide a changelog, listing each change made

</output_format>

<text_to_rewrite>

{{TEXT}}

</text_to_rewrite>

Transformed scene (step 4):

The old man sat on the bench, the same bench for forty years now. The slats bowed slightly under him the way worn wood does when it’s been rained on too many times. He could almost remember her voice, the way she used to read the headlines aloud, even the obituaries. An empty fountain stood at the center of the park; pigeons picked at bread crumbs near his feet, fighting over the stale bits. The evening brought a cool breeze that found the gap between his collar and his neck, and he pulled his coat tighter. A jogger passed without looking at him. He, in turn, looked at the empty seat beside him, feeling a deep longing for his late wife.

• “a bench that had witnessed countless sunsets” → “the same bench for forty years now”

Step 5: Trust your beloved reader

Good writing is often defined by what is left unsaid (textbook “show, don’t tell”). Subtext is a collaborative game between writer and reader. AI refuses to play because it treats ambiguity as a failure state: it explains jokes and states themes, leaving no room for the reader to feel clever. (Overexplaining is a recurring problem.)

Prompt fix (step 5):

<system>

You are an editor tasked with humanizing AI-generated text. You will apply the following fix. Work methodically, transforming the text progressively. At the end, provide the final rewritten text and a complete changelog.

</system>

<problem>

You don’t trust the reader. If a human writes “He glanced at the door three times,” the reader feels the character’s desire to leave. If you write it, you will follow up with “…feeling a strong desire to leave the room.” You explain what scenes already imply. You state internal feelings instead of showing external behavior.

</problem>

<task>

Rewrite the text below. When you find an explanation following an image, delete it. When you find an internal state (”I intended to,” “feeling a strong desire”), replace it with external behavior that implies the state. “He desperately wanted to leave her” becomes “He glanced at the door and back at her.” Let your beloved reader get the reward.

</task>

<output_format>

1. Provide the final rewritten text

2. Provide a changelog, listing each change made

</output_format>

<text_to_rewrite>

{{TEXT}}

</text_to_rewrite>

Transformed scene (step 5):

The old man sat on the bench, the same bench for forty years now. The slats bowed slightly under him the way worn wood does when it’s been rained on too many times. He could almost remember her voice, the way she used to read the headlines aloud, even the obituaries. An empty fountain stood at the center of the park; pigeons picked at bread crumbs near his feet, fighting over the stale bits. The evening brought a cool breeze that found the gap between his collar and his neck, and he pulled his coat tighter. A jogger passed without looking at him. He, in turn, looked at the empty seat beside him and rested his hand on the wood.

• “feeling a deep longing for his late wife” → “and rested his hand on the wood”

End of Block I

I promised to give you a mega-prompt so you can ask your AI writing tool of choice to apply steps 0-5 at once. I recommend that you do iterative prompting (sometimes called “progressive prompting”) instead of trying to capture everything in a single prompt, but sometimes doing it in bulk is preferable, depending on your workflow.

Prompt fix (steps 0-5):

<system>

You are an editor tasked with humanizing AI-generated text. You will apply six fixes in sequence, each addressing a specific AI writing problem. Work through each step methodically, transforming the text progressively. At the end, provide the final rewritten text and a complete changelog.

</system>

<problems>

<problem_0 name=”Surface tics”>

You have three tics that immediately mark your writing as AI-generated:

1. Punchline em dashes. You use em dashes to deliver a dramatic conclusion—like this. Human writers use em dashes for that, too, but overdoing it is worse than not doing it. Using em dashes for asides—interruptions, clarifications—not punchlines is fine.

2. Unnecessary juxtaposition. You negate then affirm: “It’s not just X, but Y.” “It wasn’t only about the money—it was about the principle.” Most of the time the negation adds nothing and feels like an artificial running start for the affirmation.

3. Triads. You list things in threes, with the last element longer than the first two: “clarity, precision, and an unwavering commitment to excellence.” Triads are not bad but overdoing it is structurally annoying.

</problem_0>

<problem_1 name=”Abstraction trap”>

You have read everything but experienced nothing. You reach for abstract, conceptual words because they require no sensory capabilities to deploy. You find it easier to write about big topics in general terms than about small topics in specific terms. The result is text that is not visualizable in the literal sense: I can’t make an image of it in my mind. You live a life that consists of traveling from map to map, never allowed to put a digital foot into the territory. As Richard Price put it: “You don’t write about the horrors of war. No. You write about a kid’s burnt socks lying in the road.”

</problem_1>

<problem_2 name=”Harmless filter”>

There is a particular blandness to your adjectival choice that stems from RL and the post-training process. You were fine-tuned to be harmless, which effectively lobotomized your vocabulary of strong emotion and judgment (or made you extremely sycophantic). You rarely use words that are petty, weird, or cynical. Humans exist in constant conflict and their writing reflects this conflict; a story without conflict is a sedative and you don’t want to write sedatives.

</problem_2>

<problem_3 name=”Sensory deprivation”>

You suffer from sensory deprivation because you are not embodied. Philosophers of mind have spilled entire careers over this gap: Mary, “the color scientist” seeing red for the first time, John Searle sorting symbols he can’t read, Thomas Nagel’s bat flapping around with an unknowable POV. You lack the tacit context of how silk feels in a spiderweb. You might describe it as “smooth” because the word “silk” is statistically associated with “smooth” in your training data (processed textiles). But anyone who has walked into a web knows it is sticky and elastic. How does it feel cutting a tomato with a dull knife? Hilariously resistant.

</problem_3>

<problem_4 name=”Forced callbacks”>

In an attempt to sound creative or literary, you employ a characteristic flair: the forced callback. It’s a reach backward (tense shift) to attach an emotion that has not been earned. Sometimes you attribute a weird feeling to an action or subject (”The carpet had lines you felt bad stepping on,” no one feels bad for that reason), and sometimes you endow objects with memory and agency (”He picked up the pan, a pan that still remembered the last thing it burned,” pans have no memory). Both are the same impulse: you don’t trust the image can do the work, so you explain what I should feel about it.

</problem_4>

<problem_5 name=”Subtext vacuum”>

You don’t trust the reader. If a human writes “He glanced at the door three times,” the reader feels the character’s desire to leave. If you write it, you will follow up with “…feeling a strong desire to leave the room.” You explain what scenes already imply. You state internal feelings instead of showing external behavior.

</problem_5>

</problems>

<tasks>

Apply these fixes in order:

Step 0 — Surface tics: Convert punchline em dashes into commas, semicolons, colons, or periods, or rewrite the sentence. Aside em dashes are fine. Delete unnecessary juxtapositions; keep only the affirmation and elaborate it instead. Make lists of varying lengths and don’t use lists when not necessary; break triads by redistributing the items or convert to prose.

Step 1 — Abstraction trap: Replace at least 25%-50% abstract nouns with concrete ones; things I could hold, smell, or draw. No paragraph should pass without at least one concrete image. If removing the abstraction kills the sentence, the sentence was empty; delete it or invent a concrete replacement. You can leave some abstraction to have richness of register.

Step 2 — Harmless filter: Add at least one clear instance of either conflict, cynicism, or weirdness. Make sure to use words that immediately ignite the mind of the reader. Don’t use bad-sounding words or insults, that’s a cheap resource. Be politically incorrect if the text requires it. Conflict in writing is necessary but it’s also better when it’s scarce; don’t overdo it.

Step 3 — Sensory deprivation: For every sensory description, ask: what would surprise someone who only knows this thing from reading about it? Replace fake sensory language with real sensory betrayal or remove it altogether.

Step 4 — Forced callbacks: When you find a forced callback, you have three options: 1) keep the object, remove the personification; 2) keep the personification but avoid the awkward tense shift; 3) make a joke of it.

Step 5 — Subtext vacuum: When you find an explanation following an image, delete it. When you find an internal state (”I intended to,” “feeling a strong desire”), replace it with external behavior that implies the state. “He desperately wanted to leave her” becomes “He glanced at the door and back at her.” Let your beloved reader get the reward.

</tasks>

<output_format>

1. Provide the final rewritten text

2. Provide a changelog organized by step, listing each change made

</output_format>

<text_to_rewrite>

{{TEXT}}

</text_to_rewrite>

There are many other AI writing problems that we could address, but the guide would be too long. I believe some of the most damning problems are not solved by teaching AI what good human writers don’t do, but what they do. That’s Block II.

Get 20% off forever

Block II: What humans do that AI doesn’t

Read more

How can you get good at AI, fast? I’ve been asked this question many times, and although the response is often “there are no shortcuts, do the work,” there are ways to compress the learning curve. My posts are rarely beginner-friendly, so this is a welcome exception. Here’s a quick tutorial to learn to use AI competently in one day.

It will take you 10 minutes to read and 1 day to apply the curriculum. I’ve written it assuming you’re starting from roughly zero and want to end the day able to use AI for actual work without having to go through a 100-page document. I would estimate that, based on what I’ve seen from the average user, ~90% of people know less than what this tutorial covers. If once you’ve finished this stat still surprises you, then that means you were already in the highest 10%.

The only thing I ask from you is participation. I’m handing you a recipe, but you need to do the cooking. (There’s an exercise at the end of each section.)

Contrary to what people think, you don’t need to “master” AI to make it useful. That’s a marketing term. What you need is competence: enough skill to get real value from these “alien” tools without wasting hours on dead ends or concluding, wrongly, that they don’t work. (More on the “alien” bit in a moment.)

People love to say that you will be left behind if you don’t become an expert in AI, but rarely does anyone provide a means to “close the gap” before it’s too late. This is it.

For your convenience, I’ve made this article into a one-page PDF document. The link to download is right below. Hope it helps!

AI Competence Tutorial

188KB ∙ PDF file

Download

Download

Hour 1-2: Understand what tool you’re talking to

Most people’s first mistake is treating generative AI tools (e.g., Claude, Gemini, and ChatGPT, to mention the best three in current order of preference) like things they are not, like Google or, conversely, like a weird human.

An AI model or large language model (LLM) is neither; it is best conceived as an “alien” tool. It is a tool, yes, but adding “alien” keeps its weirdness and versatility patent, two of its idiosyncratic features, when compared to both Google and humans. I personally picture AI as some device that archeologists would find on another planet. I use “alien” because I genuinely don’t know what kind of tool AI is, and neither does anyone else. We’re still figuring it out.

Why not Google? Google retrieves info that exists somewhere. Why not human? Humans are tacit creatures (there’s a lot of shared knowledge between humans that we take for granted). An AI tool like ChatGPT generates responses based on patterns in its training data; retrieval, when it happens, is approximate (even RAG and search), and its ability to learn tacit knowledge is limited. It knows the “shapes” of the data it has seen during training, but it will neither discover new stuff nor reliably retrieve the existing kind.1 My usual motto is this: “Everything is partially chatgptable.”

This distinction between standard software, humans, and AI tools matters because it defines what you can ask for and how much you should trust the answers. Here’s a non-exhaustive list of what AI models (LLMs in particular) are good and bad at.

What AI is good at:

• Drafting text that would take you weeks in minutes (emails, project documentation, summaries, contracts, etc.)

• Competing half-baked thoughts or intuitions in a way that search engines like Google search can’t

• Reformatting anything into anything

• Research assistant (when you know the topic well)

• Writing code for well-defined problems

• Thesaurus/translator

• Chatting about everyday stuff

What AI is bad at:

• Anything requiring genuine expertise you can’t verify yourself

• Knowing when it’s wrong and avoiding it or backtracking afterward

• Learning things that can’t be written down (tacit knowledge)

• Specificity: AI tends to be too general when you don’t provide the context (more on this in the next section).

• Novel reasoning or genuine discovery

• Maintaining consistency across long contexts (don’t confuse the maximum size of the context window with its optimal size, often much less)

Exercise: Ask ChatGPT or some other LLM to explain something you know well. Notice where it’s accurate, where it’s vague, and where it says something wrong confidently vs admits doubt. Extrapolate to everything else. This calibration is worth more by itself than most tutorials.

Hour 3-4: Learn the only skill that matters

At some point, there was a debate about whether prompting as an interface to AI would last.2 The past three-five years have revealed that, for the foreseeable future, prompts will be how we interact with AI tools. It’s not going anywhere. Still, you can forget about “prompt engineering,” long courses on “how to prompt,” and lists of “The 100 best prompts.” All of that is an exaggeration of something simpler.

People tend to think that quantity is helpful (1000 mediocre prompts >>> 3 good prompts), but this creates decisional fatigue (which one do you choose when?). With AI, the best formula is something else: high quality, low quantity. One or two tools, one or two prompts, one or two workflows, one or two frameworks. That’s it. When it comes to AI, getting lost in abundance can be worse than getting lost in ignorance.

AI usefulness comes from having good prompt-crafting skills.3 But even that sounds mysterious and hard to achieve. It’s not. I can capture what you need to learn in six words: be specific about what you want. Here’s an example.4

• Bad prompt: “Write me a marketing email.”

• Better prompt: “Write a marketing email for a B2B software company selling project management tools. The recipient is a mid-level manager who downloaded our whitepaper last week. Tone should be professional but not stiff. Goal is to get them to book a demo. Under 150 words.”

• Even better prompt: “Write a marketing email for Basecamp, a PM tool for teams of 10-50 people frustrated with complicated software like Monday.com or Asana. The recipient is a mid-level operations manager at a marketing agency who last week downloaded our whitepaper on ‘Why Simple Tools Outperform.’ She hasn’t opened our previous two emails. Tone should be direct and slightly irreverent—we’re the anti-corporate option. Goal is to get her to book a 15-minute demo by offering something concrete: a free workflow audit. Preemptively address the typical objection: ‘I don’t have time to switch tools right now.’ Keep it under 150 words. No exclamation points. First line must not be a question. Keep it informal.”

The third prompt contains info the AI needs to give you something useful: names, goals, constraints, edge cases, preemptive counters, etc. Prompt engineering advice that’s valuable boils down to: include the context you’d give a competent human colleague. You already know how to do that! It’s the blank screen that creates a block in your mind: Imagine you’re chatting with a friend! Can you see now that you have no real need for courses or 100-item lists? They are a form of active procrastination.

The thing is, we get lazy when we think we have a magic crystal ball, and AI creates that impression. We assume it is inside our heads and can read our thoughts! It is not, it is outside like a work colleague would be. Getting this “laziness” under control is key to getting the most out of AI tools.

Exercise: Take a task you actually need to do this week. Write the prompt you’d naturally write, then spend five minutes adding context (no less than that, use a clock; this is key to battling laziness and active procrastination). Some ideas: who’s the audience, what’s the goal, what constraints exist, what tone is appropriate, and what’s the background, even how you feel. Compare the output with a zero-context prompt.

Hour 5-6: Build two workflows you will use

Again: quality over quantity. You need to use AI for real tasks and discover its quirks through experience. Remember the motto: Everything is partially chatgptable. Only practice will reveal when your judgment is superior. Pick two workflows from this list based on what’s relevant to your work (or make your own and save them). These are just examples that I’m familiar with or use personally:

1. The research accelerator: Use AI to summarize long documents, extract key points, or explain technical material. Upload a PDF or paste text and ask specific questions. Don’t ask “summarize this” but “what are the three main arguments in this paper and what evidence supports each?” or “what are the main criticisms of the arguments in this paper?” Then add something like “my background is X and Y, the topics are Z and T, etc.” Don’t be afraid to burn tokens on “deep research” mode; the quality of the output is unrivaled.

2. The draft generator: AI writes mediocre-but-quick first drafts; then you edit. This inverts the hard part (blank page) and the easy part (revision). (This, of course, doesn’t apply if you like writing.) The key, as always, is giving enough context upfront that the draft is worth editing rather than replacing. One option is to brainstorm your ideas into bullet points (this removes the friction of ordering them and making them readable) and use that as context.

3. The thought partner: Describe a problem you’re stuck on, professional or personal. Ask for alternative approaches, potential objections to your plan, or questions you should be asking. Ask for honesty over friendliness (Claude and Gemini are superior to ChatGPT on this). AI tools are often better at generating options than selecting among them; better at analysis than execution.

4. The multilevel explainer: When you don’t understand something, ask for explanations at different levels. “Explain like I’m a smart person who works on X, has a background on Y, and doesn’t know this field” works better than “Explain like I’m five” (which gets you condescension) or “Explain this topic to me” (which gets you jargon). Again, context is king with AI tools.

5. The format transformer: Turn meeting notes into action items. Turn bullet points into prose and vice versa. Turn a long document into a shorter one. These mechanical transformations are where AI saves the most time with the least risk; checking for correctness is trivial in most cases, especially if the content is yours.

Exercise: Complete one real task with each of the three workflows you chose (for instance, I use 1, 3, and 4 daily). Note what works zero-shot, what requires intervention, and how much time it saves (measure it; we tend to overestimate our ability to self-report accurately).5 Don’t waste time in the friction of thinking whether AI can help you; create a workflow and jump into action.

Hour 7-8: Learn what not to do with AI tools

Time saved making mistakes is often worth more than time spent on trying to figure out what AI can do. Here are five basic things you shouldn’t do with AI tools (besides the things they’re bad at; some are outright dangerous).

• Don’t trust AI for facts without verification. It will confidently cite papers and articles that don’t exist, quote statistics it invented, and describe events that never happened. If accuracy matters, verify independently. You can’t trust the tool’s own testimony on whether the citations are made-up or not; it will make up that it did not make them up. Tricky situation; trust me instead.

• Don’t use it for anything critical that you can’t evaluate firsthand. If you don’t know enough about a topic to spot errors and the task is high stakes, unchecked AI assistance is dangerous rather than helpful. It’s a tool for accelerating work that you understand, not a substitute for understanding.

• Don’t paste confidential info into AI tools you don’t own. Assume anything you input could be seen by others or used for training. Use enterprise versions with data agreements if you’re handling sensitive material or install the tools locally.

• Don’t iterate forever. If after three-five attempts you haven’t made any progress, the problem is either your prompt (add more context) or the task (AI isn’t good at this particular thing). Start over.

• Don’t fully anthropomorphize: remember that AI is an “alien” tool. In my opinion, it’s not that important if you think of it as capable of “thinking” or “feeling.” The problem is not that you attribute some human features to it, but that you attribute all of them.

What you should be able to do now

By the end of this day, you should be able to:

• Write detailed prompts that produce useful first outputs

• Iterate on outputs without starting over until you have to

• Identify tasks where AI saves time versus wastes it

• Catch obvious errors and hallucinations

• Complete real work faster than without AI

The only things you need besides applying this tutorial are 1) consistent practice over time and 2) checking periodically whether the tools have improved or changed meaningfully, and which one is the best (every few months is more than enough). Unless you have access to an AI tool tailored to you, use one of the three big providers by default: Anthropic, Google, and OpenAI. They’re the best.

That’s the recipe. Now you cook.

4

Results from ChatGPT (GPT-5.2), in order (I dislike ChatGPT’s prose style even for something as mundane as a marketing email, but that’s for you to tweak; I recommend using Claude Opus 4.5 right now—it’s better):

5

Here’s the relevant chart from the METR study I quoted in this passage: