[Journey] Creating Good Books With AI

I will focus on the quantity and going to upload more, hopefully I will get some profits
You're right, Etsy is better, you making good on POD mugs bro ?
I plan to make some journeys with POD(tshirt) on Etsy
Mugs are my full time income. KDP is a side interest at the moment.

I have another product I am looking to launch later this year which is POD but if it works I will transition to holding stock. Thr market up is ridiculous compared to POD on this one.

Never done the tshirt thing. Are your books and Etsy linked or completely separate
 
My other books arrived and I think I have found an author I really enjoy reading Erik Larson does this creative non fiction. The Devil in the White City was a really interesting read. Not sure if there’s any practical implications for writing Good books with AI.

One thing I am going to try comes from two different sources. One guy whose name I forget talked in a podcast about every page needs to earn its place in the book. I think that’s gold. I am not sure on how to implement that just yet but it feels right.

I think in someways this may tie into what Dean Koontz does with his heavy editing and reworking of every page before he starts the next. He’s a fiction writer so it doesn’t quite translate but I think that’s a way of improving a chapter.
 
I think I had a bit of a eureka moment today when YouTube played a random video on a phd level paper.

It got me thinking about the process of writing and so I went across to several AI chatbots to see what they would say.

The consensus was that to write a book or chapter AI has to:

1. establish the book premise,
2. understand the target audience
3. address the book purpose
4. research the subject matter
5. select what needs to be include,
6. develop a chapter skeleton
7. select a voice and tone
8. write the the chapter


The end result is that I think asking AI to write a good book or even a chapter is too much in one pass.

I am starting to wonder whether one brain for one task might be a better option. Ask chat got to research the subject first then write based on that. I guess in someways this is closer to what a human would do.

I don’t know but it’s an exciting thought and feels like my first move forward
 
After my last post I wanted to test whether breaking the writing process down into separate tasks helped produce noticeably better output.

For this experiment I want to do a single variable test.

The first experiment which we have done before was simply a one and done prompt Prompt: One-shot “Write a 1,500-word book chapter on saying no for a book on “people pleasing”

Then a simple structured Multi-Brain Sequence split into two prompts:

Prompt 1: Research and summarize key points, statistics, examples, and common pitfalls on the topic.

Prompt 2: Using only the research output above, write a 1,500-word book chapter on saying no for a book on “people pleasing”

So the second prompt in experiment 2 is deliberately the same as in experiment 1 but with the confined of using the research in prompt 1 of experiment 2.

I ran everything in incognito to avoid any carry-over from previous chats.

The multi-brain approach produce a much better book chapter. Notably so.

In case this was a fluke I repeated the experiment a further 11 times and consistently the multi brain method produces higher quality results.

The multi brain method versions felt like they flowed better were more specific (even directly referencing some of the research)

Now to be honest it’s far from perfect. The voice issues are still there it still reads like AI but on a side by side read there’s a clear winner.

I feel like this is a much needed win. Having listen to so many people on YouTube and social media on how easy KDP is I honestly thought I would have at least a draft of a book by now.

For now I am going to put a tick next to multi brain. Next I’m going to expand this into a longer chain and see if there comes a week point. Context window may come a problem but we will see.

Has anyone else experimented with prompt chaining like this would love to hear your experiences.
 
So I think I have an idea to test based on multibrain and it goes back to the books I bought earlier.

One book I liked was Brainfluence and I have kept going back to it. Perhaps to write a good book I shouldn’t be writing a traditional non fiction but rather a modular or listicle style book.

Use the multi brain technique on each chapter and end up with x amount of smaller chapters that each cover one topic.

To my mind that gets the best out of AI minimises the issue with context windows and should produce a book of substance.
 
I've written some long form books and Ai does best when its given a structure to follow, even then it needs heavy refining afterwards. Took me a few hours to get like 120 pages even readable to a normal person. And then a few more days in editing and changing sentences to get it somewhat okay. You also need to distill a writing style down into a prompt and give it examples to write like otherwise it just drifts into repetitive nonsense. Good luck!
 
I've written some long form books and Ai does best when it’s given a structure to follow, even then it needs heavy refining afterwards. Took me a few hours to get like 120 pages even readable to a normal person. And then a few more days in editing and changing sentences to get it somewhat okay. You also need to distill a writing style down into a prompt and give it examples to write like otherwise it just drifts into repetitive nonsense. Good luck!

I have noticed in my AI experiments so far that the flow of sentences doesn’t follow a logical or sequential pattern. For those into theology it’s a Markian sandwich. Here’s A here’s b and here’s A. Interesting and not obvious choice by the machine.

I do have writing style to complete later on but I agree it’s vital and not just putting in the style of or inspired by x. I like the idea of al lay creating an avatar for the author beyond just words they use. This is for further down the line

What did you end up doing with your book(s)?
 
I'm not sure AI can write a viral book because it doesn't have a unique character (yet). However, you should write the book for yourself. If you can't stop reading it, then it's fine.
 
thank you I appreciate your response and I largely agree with what you are saying - there is a ton of slop out there.

I can well believe 99% of people using AI to publish are slapping a generic prompt into ChatGPT, and dumping an unformatted PDF straight onto Amazon expecting to make millions overnight.

But just because the market is flooded with bad AI books doesn’t mean you shouldn’t use AI. It is like saying you shouldn't build a niche website because there are tens of thousands of crappy WordPress blogs out there.

But that’s exactly why I started this journey.

I’m not doing a "one-click" spam method.

I want to see what AI is actually capable of. Through trial and error can I get AI to write a good.

If I can’t I won’t publish anything.
It means that you edit the book manually yourself right? You can't completely let AI write the whole book without a human touch. Keep us updated, Best of luck!
 
It means that you edit the book manually yourself right? You can't completely let AI write the whole book without a human touch. Keep us updated, Best of luck!

That’s an interesting question and I am not sure I have an answer at the moment. I have been primarily focused on creation up until this point.

Thinking off the top of my head I envisage that there will be some AI assisted editing followed by human input.

For the ai assisted I can see the multibrain concept working. Having an AI as a line editor, a development editor and later a proofreader.

The amount of human input will probably depend on the final creation methodology.

So if I end up going with something similar to Koontz method then there is a lot of drafts and redrafting going on. Issue that would be picked up in a normal first edit will have been tackled before hand.

If I go with first draft first kind of method then there will be a heavier editing process. It’s on this I could see multibrain then human editing being needed a lot more.
 
That’s an interesting question and I am not sure I have an answer at the moment. I have been primarily focused on creation up until this point.

Thinking off the top of my head I envisage that there will be some AI assisted editing followed by human input.

For the ai assisted I can see the multibrain concept working. Having an AI as a line editor, a development editor and later a proofreader.

The amount of human input will probably depend on the final creation methodology.

So if I end up going with something similar to Koontz method then there is a lot of drafts and redrafting going on. Issue that would be picked up in a normal first edit will have been tackled before hand.

If I go with first draft first kind of method then there will be a heavier editing process. It’s on this I could see multibrain then human editing being needed a lot more.
Alright, there's no problem. You can do whatever you want provided that you get the sales volume that you want. Best of luck!
 
Alright, there's no problem. You can do whatever you want provided that you get the sales volume that you want. Best of luck!
Thanks as I say I do t know yet with regards to editing. Open to suggestions and ideas.
 
I ran the modular / listicle test I mentioned last time by doing the following:
  • Chose a narrow self-help niche
  • Used the multi-brain method (research pass → writing pass) on each individual section
  • Aimed for short, self-contained chapters/sections so I could avoid context window issues and keep quality higher on each piece
  • Fact checked with different AI’s

It didn’t work as well as I’d hoped.

The individual sections in isolation were okay. They were clearer and more specific than my earlier one-shot attempts which I still see as a win and could be the basis of something. when I put the chapters together and read the whole thing as a book, several problems became obvious:

Many sections felt padded. Paragraphs or saying anything.

The information was surface-level once they were next to each other. What looked decent on its own started felt repetitive over 100 chapters.

The book ultimately felt more like a collection of blog posts than a cohesive piece of writing. There was no sense of progression, purpose or transformation.

Even with the multi-brain approach, the tone drifted between sections. It’s like different windows had different AI personalities in it.

A lot of the “research” AI pulled was the same well-worn studies and examples that appear in every book in the niche. It lacked the sharper, more original insight I was hoping for. It doesn’t even have to be new insight but it should feel that the author has somehow lived it.

In short, the modular structure helped with some technical problems (context window, focus) but didn’t solve the deeper quality problems. It just packaged them differently.

I’m not writing the format off completely I feel like there is something there and books like Brainfluence and Cashvertising prove the format. The question is can AI do it.

It hasn’t all been a complete right off. In my experiment I noticed that even within the confines of multi brain there was still room for hallucinations.

Thanks to a random youtube video I came up with a workflow that I think may solve it thanks to some recent changes in AI.

I now aim to create a test loop on a subject I know. If I can eliminate the hallucinations before the edit that’s a big step forwards
 
So I have had a couple of experiments on the go at once. One dealing with the hallucinations and the other with the author.

This is the output and I would be interested in people’s thoughts on it:

Chapter 4
THE PARADOX OF CHOICE
When Abundance Becomes a Weapon


The art of being wise is the art of knowing what to overlook.
—William James


I can admit it freely now. For years I was the kind of shopper who stood frozen in the cereal aisle, comparing the relative merits of twenty-seven varieties of granola as if the decision would determine the course of Western civilization. I have spent longer selecting a bottle of wine than some people spend choosing a spouse. And more than once I have left a store empty-handed—not because nothing appealed to me, but because too many things did. The sheer volume of options had somehow drained the pleasure from the act of choosing itself.


This is not a personal failing unique to me. It is a modern affliction, and a highly exploitable one. We have been taught since childhood that freedom is maximized by maximizing choice. More options equal more liberty, more opportunity, more happiness. The idea feels so self-evidently true that questioning it seems almost un-American. Yet a growing body of evidence—and a great deal of quiet human misery—suggests the opposite: under many conditions, more choice does not liberate us. It paralyzes us, exhausts us, and leaves us less satisfied with whatever we finally select.


I call this the paradox of choice. And like the other weapons of influence we have examined, it is a principle that compliance professionals have learned to use with considerable skill.


The Jam Experiment and the Click of Overwhelm


The classic demonstration is almost too neat. In 2000, psychologists Sheena Iyengar and Mark Lepper set up a tasting booth in a gourmet grocery store. On some days they offered shoppers a display of twenty-four exotic jams. On other days they offered only six. The larger display attracted more attention—more people stopped to look. But when it came time to buy, the numbers reversed dramatically. Shoppers who saw the extensive array were only one-tenth as likely to purchase a jar as those who saw the limited selection.


Twenty-four choices produced interest. Six choices produced action.


The same pattern appears in far more consequential domains. When employers offer workers a large number of retirement-fund options, participation rates drop. When online dating sites present hundreds of potential partners, users become less likely to contact anyone and report lower satisfaction with the people they do meet. When physicians are given more treatment alternatives for a hypothetical patient, they become more likely to choose the status-quo option—or to choose nothing at all.


What is happening here is a form of decision paralysis that closely resembles the fixed-action patterns we saw in lower animals. Faced with an overload of options, the human decision-making system does not rise to the occasion with greater wisdom. It often simply shuts down. Click, and the “I can’t decide” tape begins to play. Whirr, and the person either walks away or seizes the first option that reduces the discomfort—frequently a suboptimal one.


Why More Feels Like Less


Several psychological mechanisms drive the paradox.


First is the sheer cognitive cost. Each additional option requires comparison, evaluation, and the mental simulation of future regret. The brain, which evolved to make rapid choices among a handful of alternatives on the savanna, treats a supermarket aisle the way it once treated a dense thicket of possible threats: with rising stress and a preference for escape.


Second is the escalation of expectations. When only three jams are available, any of them can feel like a reasonable success. When twenty-four are available, the chosen one must not merely be good; it must be the best. The gap between the chosen option and the imagined perfect option grows wider with every alternative added to the set. Satisfaction shrinks even as objective quality may rise.


Third is the intensification of regret and self-blame. With few options, a disappointing outcome can be attributed to limited alternatives. With many options, the disappointment is more readily attributed to one’s own poor judgment. “I should have known better” becomes the dominant post-decision narrative. This is why people who face extensive choice often report lower happiness with their selections even when those selections are, by objective measures, superior.


There is also a social dimension. In a culture that celebrates choice as a moral good, admitting that one is overwhelmed by options can feel like a personal failure. We therefore continue to demand more alternatives even as those alternatives make us miserable—a consistency trap of the purest kind.


How the Principle Is Used Against Us


Compliance professionals have not missed this vulnerability. Consider the modern car dealership. The salesperson who begins by showing you every possible trim level, color, and option package is not necessarily trying to be thorough. He is, in many cases, deliberately inducing choice overload so that when he later “simplifies” the decision by recommending a particular package, you experience relief and gratitude. The relief itself becomes a form of reciprocity: he has rescued you from the paralysis he helped create.


Online retailers employ the same tactic with “filter” systems that first overwhelm and then appear to rescue. Streaming services that present endless rows of titles often see higher engagement when they introduce curated “Top 10” lists or “Because you watched…” recommendations that artificially constrain the field. Dating apps that once competed on the size of their databases now compete on the quality of their matching algorithms—algorithms that, not coincidentally, reduce the user’s sense of infinite choice.


Even charitable organizations have learned the lesson. A fundraising letter that asks donors to choose among twelve different programs often underperforms a letter that presents a single, clear request. The organization that appears to offer less choice frequently receives more money.


The Scarcity Connection


Notice how neatly the paradox of choice interacts with the scarcity principle we examined earlier. When options are genuinely limited, scarcity increases perceived value. When options are artificially abundant and then selectively restricted, the restriction itself becomes a powerful influence tool. The marketer who first floods you with possibilities and then says, “But this one is available only today,” is playing both sides of the same psychological coin. Abundance creates the anxiety; scarcity provides the relief—and the sale.


Living with the Paradox


If the research is correct, the solution is not to eliminate choice. Some degree of autonomy is essential to well-being. The solution is to become strategic about the conditions under which we allow choice to expand.


One effective defense is pre-commitment to decision rules. Before entering the environment of abundance, decide in advance the criteria that matter most and the maximum number of options you will seriously consider. Another is the deliberate cultivation of “good enough” standards rather than maximizing ones. The maximizer who insists on the absolute best is far more vulnerable to the paradox than the satisficer who seeks only an option that meets a predetermined threshold of quality.


There is also value in outsourcing certain classes of decisions. The person who wears the same style of clothing every day, or who always orders the house specialty, is not lacking in freedom. He has simply recognized that some freedoms are more costly than they are worth.


Finally, we can learn to treat the feeling of overwhelm itself as a signal. When the number of options begins to produce anxiety rather than pleasure, that anxiety is information. It is the mind’s way of saying that the current choice environment has crossed from liberating to coercive.


A Closing Observation


I still catch myself lingering too long in the cereal aisle from time to time. Old habits die hard. But I have learned to recognize the moment when the pleasant hum of possibility turns into the low-grade panic of excess. At that point I apply a simple rule I now teach my students: when the options begin to multiply beyond the point of comfort, the most empowering choice is often the choice to choose less.


In an age that relentlessly sells us the idea that more is always better, the ability to impose intelligent limits on our own freedom may be one of the last remaining acts of genuine autonomy. And like all the other weapons of influence, it works best when we understand it well enough to decide—consciously—when to hand it to others, and when to keep it firmly in our own hands.

————

There is still some AIisms and I haven’t ran it through any kind of detector but I like how it reads feels very natural.
 
I have settled on a two AI set up. Ultimately Claude has the best prose and can be changed with the stylometry instruction. ChatGPT just produced one line paragraphs which read well but were horrible on the page.

I have added an AI pass based on common trends and this helps clean the language up. One thing I am struggling with is word counts. AI just can’t count and consistently undershoots what should be created.

I am happy with how things have moved on and learned a lot from YouTube channels like future fiction academy. It feels for the first time that something is coming together.
 
Back
Top