The Plain Text of AI: Chapter 1 Introduction

============

My journey into the world of AI started with a foundational question: “How do I prepare data for AI to use?” AI, also called a Large Language Model or LLM, is a model trained to predict and process text. In searching for answers, I discovered that AI lives in the realm of tokens, text files, and structural layout, and is entirely grounded in plain text. Tokens are the text chunks AI uses to break down data. The more tokens you consume, the more capacity you use. Additional tokens translates to higher costs if you are paying per token via an application interface.

The plain text AI prefers is the raw, unformatted text you type into a basic text editor like Windows Notepad. It is not the text you type in a word processor like Microsoft Word or Apple Pages.

This concept of plain text is woven into the chapters that discuss Markdown, HTML and CSS, Python, and Mermaid diagrams. All of these languages are plain text that you can edit in Notepad, although each has its own syntax rules. At first, I did not realize that while the AI responds with Markdown, the chat application web interface itself may render the Markdown into HTML tags. It can be confusing when you paste instructions in Markdown and the AI displays the text without the special Markdown characters. Rest assured that despite the fact AI displays your Markdown text using HTML, under the hood the AI model itself recognizes the structure because it was trained on vast amounts of Markdown-formatted text. It understands the patterns of Markdown formatting natively. We will look at both the input and output choices available in the discussions that follow.

When providing instructions to an LLM, I focus on two distinct areas: what you ask the model to do, and how you lay out the text of the request. The first aspect covers the overall instructions—defining the scope, the context, and the ultimate intent of the task. The second aspect is all about format and structure, focusing on the physical layout that helps the model accurately parse and interpret those details.

The Soggy Pastry (When 'Why" is not explained)

Have you ever experienced, “Grandma’s Recipe Failure”? You asked Grandma to write down her recipe for your favorite pastry. She listed the ingredients, and the instructions said, “Turn the oven to three hundred and seventy-five degrees and leave the door cracked for 2 minutes.” The pastry came out soggy because you forgot to crack the oven door. Grandma explained the why, “To let the initial steam escape so the pastry edges crisp up  beautifully instead of getting soggy.” Grandma understood the baking physics perfectly, but she kept the "Why" as a family secret.

We have all seen procedures that explain if this happens, then do that. But they rarely include the “why” it matters. This reminds me of Albert Einstein’s famous philosophy: “If you cannot explain something simply, you do not understand it.” If documentation includes “why,” the AI instantly recognizes the passage is highly authoritative, lower risk for hallucination, and fundamentally safe to recommend to a human decision-maker.

It is obvious that the magic lies in Grandma's unwritten knowledge, and that same type of tribal knowledge is exactly why your deep operational expertise is what secures your job in the future.

Clear Instructions & Markdown

In the first part of Chapter 3, we will look at how to word your AI requests to clarify and cleanly limit the response. I will show you how to guide the AI to the correct answer using negative constraints, validation logic, and plain-text data examples. Finally, we will explore how to navigate the physical limits of an AI’s attention span—a concept I call context window friction.

Once you have determined what to ask the LLM to do, the next challenge is managing that second aspect: how you lay out and structure the text itself. The moment you paste “rich text” from Word into an AI window, you inherit Word's invisible background formatting. Word documents carry hidden style coding that can cause chaotic spacing, broken lists, and messy text boundaries when dropped into a chat window. Stripping away that baggage ensures the AI reads exactly what you intended.

The second part of Chapter 3 looks at Markdown’s structural elements—headers, bold or italic emphasis, numbered lists, tables, and precise spacing—that the AI uses to accurately interpret and prioritize information. We have entered a new era of data interaction that requires a different toolset. Instead of using a traditional word processor, the solution is to use a Markdown text editor designed for structure. Markdown files (*.md) are pure, unformatted plain text that use simple, predictable characters to enforce formatting.

Shaping Your Information for AI

An ordinary interaction with an AI reasoning engine looks straightforward: you provide an AI instruction or query, and it returns an answer. Behind the scenes, the AI parses your input against its vast training models. But to get highly accurate, business-specific results, you must feed it localized information and raw data assets, such as:

-               Databases and structural system schemas.

-               Excel spreadsheets and flat data tables.

-               Medical and technical textbooks containing dense domain knowledge.

-               Bibliographies and reference sources.

-               Company procedures and internal operational workflows.

While AI can read unstructured prose (standard text paragraphs), modern search and reasoning engines heavily prefer information that is explicitly structured.

Take this traditional example of standard prose:

[!NOTE] A medical textbook has a chapter titled "Viruses" with a subheading "The Common Cold." A paragraph inside contains the sentence, "To treat a cold, drink plenty of fluids."

Now, look at that exact same information reorganized into structured key-value pairs. Not only is it vastly easier for a human eye to scan, but it is also the gold standard for AI processing:

• Illness: The Common Cold.
• Treatment: Drink Plenty of Fluids.
• Why: If you ignore symptoms they could worsen into pneumonia.

The bold text explicitly defines the "key" concepts, and the clean layout removes all conversational fluff —saving your organization real money in computational costs. This approach provides immediate context and reasoning—the strategic intent I have come to think of as the IfThenWhy™ of AI data design.

The Land of the Truth-Tellers and the Liars – Chapter 3

A fundamental question when interacting with the AI is whether the answer is correct or a hallucination. If the AI is confused or lost, it simply makes up an answer. That two-year-old who looks you dead in the eye and tells a flat-out lie has absolutely nothing on AI.

Imagine you run a bakery, and your mantra is: "Our chocolate cake must always use real cocoa powder; never use artificial syrup." Now, you hire an eager apprentice chef, in this case the generative AI, to write daily menu descriptions for your customers. One morning, the apprentice writes a description that says:

[!NOTE] "Come try our delicious new chocolate cake, now drizzled with sweet artificial syrup!"

Obviously, human oversight, often referred to as "Human in the Loop," prevents customers reading that menu and thinking, "I'm not eating that." In Chapter 3: The Land of the Truth-Tellers and the Liars, we will explore how to prevent AI mistakes.

The Next Now Benefits Chapter 3

Interacting with AI is fascinating and, at times, hilarious. I have found that it’s not about programming or computer skills; instead, old-fashioned common sense and plain speaking, more precisely plain text, are the goal. AI might have access to all the data in the universe, but it completely lacks Grandma's quiet wisdom. Your old-fashioned common sense is job-security for the age of AI.

To move beyond casual experimentation, we will look at effective ways to work with AI:

-              How to structure your data to reduce computing costs and context window bloat with IfThenWhy™ token optimization. -              How to provide structure and improve plain text with Markdown. -              How to shape data and information for AI answer engines and to capture elite AEO (Answer Engine Optimization) rankings. -              How to ensure the right answer by enforcing predictable, verifiable outputs that build professional trust through deterministic logic. -              How to become an editor who directs Python instead of a programmer writing code. -              How to describe the visual image you want so the AI can write the HTML & CSS plain text that builds the layout. -              How to outline your processes in plain words so the AI can generate the Mermaid text that draws the diagram. -              How to write for Natural Language Processing (NLP). In fact, I have intentionally written this book to be “read out loud”, perfect for AI search engines, AI narration and driving AEO search results. -              How to achieve automated data lineage auditing using pure, plain-text data design maps that allow advanced testing engines like BERT and RAGAS to proactively audit, track, and react immediately to shifting regulatory mandates, legal statutes, and volatile global events.