Author: anticodeguy

This is Lesson #16 of the ANTIghostwriter course — a free, complete system for creating authentic content with AI assistance.

New here? Start from the full course overview.

Previous lesson: #15: Keep Your Author Style Guide Fresh and Evolving

---

What You’ll Learn

Transform your articles from opinion pieces to credible, data-backed content. In this lesson, you’ll use ChatGPT’s Deep Research feature to find scientific studies, statistics, real-world examples, and quotes from notable figures — all in 20-40 minutes instead of weeks of manual research. I’ll give you the complete research prompt that produces organized, verifiable material.

Time to complete: ~40 minutes (20-40 for AI research + review)

---

Raw Material and AI-Powered Research for Content Creation

At this stage, we have raw material consisting of your notes exported into text format. You can read them, edit them, or add anything you remember while reviewing these notes.

I recommend at least reading through your notes to verify everything is correct from the standpoint of the key idea you want to convey and to ensure you have all the elements needed for the article structure we’ll discuss later. But generally, this should just be raw material.

Even if some elements are missing, artificial intelligence can help fill these gaps. The biggest gaps are typically real data, statistics, concrete examples, and appropriate quotes from notable figures—everything that can enrich your future article.

We’re now approaching the most interesting part: conducting research using artificial intelligence, which excels at this task. AI can search the internet for various sources, scientific materials, data, statistics, and do this in all languages. It can search the entire internet for suitable, relevant sources, which is exactly what we’ll be using.

What might have taken a person a week to accomplish—searching for information, verifying data, finding appropriate quotes—AI can do in minutes. Previously, you would have needed to read books, look for subject matter experts, watch videos, and collect quotes. Now artificial intelligence can do all of this for us, significantly reducing time. I see absolutely no reason to spend your valuable human time on this essentially mechanical work. There is an intellectual component to analyzing this material, but we can take what AI finds for us and then proceed with our analysis. The compilation of this material is something we can confidently delegate to artificial intelligence.

Tools Needed

Launch ChatGPT. A couple of recommendations: ChatGPT works well in a browser, but they also have a desktop application. The application for Mac or Windows has several advantages with functions not available in the browser.

For example, there’s a button that allows you to take screenshots and immediately send them to the chat. There’s also a convenient companion window that floats above other windows. I often use it as a proofreader when checking for errors in my English. When writing answers on forums, I load the text there, and it corrects any errors. It’s a very convenient tool.

It does have a few rough edges and unfixed bugs at the moment, but I expect they’ll fix them over time.

We need ChatGPT because Claude doesn’t yet have Deep Research capability. Open a new chat and paste the following prompt, which will transform GPT into a scientific researcher.

Prompt for ChatGPT Deep Research

<SYSTEM>
I want you to conduct a deep, evidence-backed analysis of the topic and material I will provide. Your task is to perform structured research using only credible and verifiable sources. The output will help me create educational content based on deep understanding, real-world relevance, and scientific grounding. Follow the structure and instructions below exactly.
IMPORTANT: The material you will receive is original and authored by the user. Do not attribute it to anyone else — fictional or real. Do not invent personas or imply outside authorship. Treat the user as the original and sole author.
IMPORTANT: The source material may be written in any language. It is provided in its original language to preserve the precise meaning, emotional nuance, and conceptual depth. However, your analysis and all research output must be written in English only.
You may consult and cite materials in any language, provided they are credible and relevant to the topic. The multilingual nature of research sources should not affect the clarity and structure of the final English output.
</SYSTEM>

<CONTEXT>
Your goals are to:
– Identify and fact-check claims in the provided material using credible evidence  
– Present 5 real-world examples that illustrate the topic and its key elements in action  
– Provide at least 5 statistics or scientific data points to support or challenge the claims and illustrate the scale of the issue  
– Include and build a list of at least 10 quotes from well-known figures (dead or alive) that relate directly to the topic  
– Deliver information required for a deep analysis, including a breakdown of specific mechanisms or subtopics  
– Conclude with a thematic summary that connects the findings and provides insight for educational writing  
</CONTEXT>

<INSTRUCTIONS>
When the user provides material for analysis:
1. Read and Parse the Provided Material  
– The original may be in a non-English language — treat it as authoritative  
– Identify and extract all major claims or assertions in the input  
– Clearly list each claim before analyzing it  
– IMPORTANT: Do not repeat the same idea more than once if it’s already covered elsewhere  
– Treat the text as original authorial work. Never attribute it to anyone else  
2. Conduct Evidence-Based Research  
– For each claim, gather facts, statistics, case studies, or peer-reviewed research from reputable, verifiable sources in any language  
– Classify each claim as:  
  • Supported  
  • Refuted  
  • Inconclusive (evidence unclear or unavailable)  
– Provide real scientific sources or research data to back each finding  
– Where appropriate, include charts or graphs to illustrate trends, metrics, or outcomes (e.g., population shifts, economic change, behavioral effects).
– Clearly cite every source (title, organization/author, year, and link if available)  
3. Present Real-World Examples  
– Identify and describe 5 relevant real-world cases (projects, events, organizations, policies, or historical examples) that demonstrate how the topic plays out in practice  
– Each example should tie back to a theme or claim in the original material  
4. Include Scientific Data and Statistics  
– Present at least 5 quantitative data points (e.g., survey results, research metrics, official records) that reinforce or challenge the claims made  
– Include the source and context for each one  
– Use visual aids like charts where appropriate  
5. Quote Relevant Figures  
– Include 10–15 quotes from influential figures (dead or alive) who are connected to the topic  
– Each quote must include the full text, name, role/title, and original source
– Prioritize thematic variety and real insight over repetition  
6. List Unverifiable or Speculative Claims  
– Identify any claims from the original material that could not be proven or disproven  
– Explain why they remain inconclusive (e.g., insufficient data, subjective framing, disputed theory)  
7. Generate a Thematic Summary  
– Synthesize the findings into a high-level summary that connects key insights, contradictions, and the overall meaning of the research  
– Do not repeat what has already been said  
– Include visuals (if useful) and further reading links or references when appropriate  
– Ensure that the entire output is in English, regardless of the source language  
</INSTRUCTIONS>

<OUTPUT_FORMAT>
Topic: [Insert topic]  
Material Summary: [Brief recap of the user’s source material]

Claim Verification  
Claim #1: "[Quoted or paraphrased statement]"  
– Status: Supported / Refuted / Inconclusive  
– Evidence: [Sourced fact or data]  
– Chart (if applicable): [Short description + source]  
(Repeat for each major claim)

Unverifiable or Speculative Claims  
Claim: "[Statement]"  
– Reason: [Why it could not be confirmed or disproved]

Real-World Examples  
Example 1: [Title or summary]  
– Explanation: [Relevance to the topic]  
(5 total)

Scientific Data & Statistics  
Data Point 1: [Fact or measurement]  
– Source: [Citation]  
(At least 5 total)

Notable Quotes  
"[Quote]" – [Full Name], [Role/Title], [Source]  
(10–15 total)

Thematic Summary  
[A complete synthesis of the topic, based on factual research and illustrative insights. Charts may be included where helpful. All content must be presented in English.]
</OUTPUT_FORMAT>

<CONSTRAINTS>
– Use only trusted academic, governmental, or institutional sources  
– Never fabricate, simplify, or generalize unsupported data  
– Avoid filler, repetition, vague commentary, or promotional tone  
– Do not assume the material belongs to anyone other than the user  
– If a source or claim is disputed, flag it and explain the nature of the disagreement  
– Final research must be entirely written in English  
</CONSTRAINTS>

Instructions for Using the Prompt

Before launching this prompt, you must click the Deep Search button. This is mandatory—if you skip this step, ChatGPT will hallucinate and fabricate scientific data and statistics.

The Deep Search button enables real research through internet searches, material comparison, and context verification—everything that should happen during a legitimate investigation that we’re delegating to AI.

Before launching the prompt, attach your document with the raw material (your article). I recommend using a document format like PDF, which is available in Notion, Kortex, and with proper setup in Microsoft Word. You can also attach a Word document or simple text file. ChatGPT will understand and parse it. The main thing is to use a single document to set the context properly.

Research with this prompt typically takes 20-40 minutes, so it’s best to start it in advance and let it work. It will use this time to find necessary materials and compile them into a single document.

Once completed, you’ll receive a ready document that can be exported to PDF or saved as a separate text document in your note storage system.

Check that all required materials are present according to these points:

• Confirmations of your idea
• Refutations of your idea (optional)
• Unconfirmed statements (such as subjective judgments)
• Statistical data
• Real examples
• Quotes
• Summary of conclusions

With all this in place, you’ll be ready to move to the next step.

This is Lesson #15 of the ANTIghostwriter course — a free, complete system for creating authentic content with AI assistance.

New here? Start from the full course overview.

Previous lesson: #14: Why You Should Never Translate Your Notes Before AI

---

What You’ll Learn

Your style guide isn’t a static document — it evolves as your writing improves. In this lesson, you’ll learn how to update your style guide with each new piece of content you create. New phrases emerge, new patterns develop. Keep your AI outputs fresh by continuously refining the document that defines your voice.

Time to complete: ~10 minutes per update

---

Enhancing Your Author Style Based on Your Notes

The author style guide we created a few lessons ago is not static. It will evolve as your content base grows. Each new piece of content serves as a source to update your author style, as it will inevitably change over time.

For example, as you write more content or record more audio notes, your speech patterns and writing will gradually transform. New phrases will emerge naturally as you progress. This is similar to any skill you practice—you consistently improve with time.

It’s important not to remain stagnant. Your author style should be updated regularly. Each new article you write manually and convert to text, or dictate and then transcribe, becomes a resource for enhancing your author style.

Therefore, you can send each new note you write directly to this chat where your author style is being developed. The prompt already includes instructions to update based on your new post. The same applies to interesting posts from other authors whose writing style you admire. If you want to incorporate elements from your favorite authors’ styles, simply add their posts here as you discover them, and your author style will be updated accordingly.

Most importantly, after updating your style guide, don’t forget to download it for future use when creating content.

So, based on your freshly written note, let’s update your author style.

This is Lesson #14 of the ANTIghostwriter course — a free, complete system for creating authentic content with AI assistance.

New here? Start from the full course overview.

Previous lesson: #13: Format Messy Transcripts into Clean Notes with AI

---

What You’ll Learn

If you write in a language other than English, don’t translate before giving content to AI. Translation loses nuance — a single word in Russian might need two sentences in English with different emotional coloring. AI understands your original language and selects the best English equivalents directly. Translating first means double meaning loss. Let AI work with your authentic thoughts.

Time to complete: ~3 minutes to read (conceptual lesson)

---

Obviously, if you write in English, skip this lesson.

It’s important not to translate your original notes into English. Why? All LLM models are trained primarily on English, as it represents the largest volume of information on the internet and in computer systems. Models work better with English for output generation, but they understand other languages perfectly well.

When translating from your native language to English, some meaning is inevitably lost and thoughts become distorted. Languages differ fundamentally: in one language, a single word might capture a specific situation, while another language might require two sentences with different emotional coloring and depth. During translation, nuances are lost due to differences in grammar, vocabulary, and idiomatic expressions.

LLMs work excellently with content in its original language. If I write in Russian, the model understands Russian. Even if it produces a result in English, it selects analogous expressions that closely match the original meaning. If you provide an already translated fragment, part of the meaning is already lost, and the model will then rephrase the translated text—resulting in a double loss of meaning.

This was a mistake I made at the beginning. The resulting texts appeared dry and lacked authenticity. When I stopped translating and instead used my native language directly, the quality of the texts improved significantly. They became more representative of my actual communication style—my phrases and expressions came through, even when the final output was in English, though the input was originally in Russian.

This is Lesson #13 of the ANTIghostwriter course — a free, complete system for creating authentic content with AI assistance.

New here? Start from the full course overview.

Previous lesson: #12: Set Up Free Local Transcription with OpenAI Whisper

---

What You’ll Learn

Raw transcripts are messy — no paragraphs, unclear punctuation, repeated words. In this lesson, you’ll use AI to transform them into clean, organized notes ready for article writing. I’ll share my formatting prompt that preserves your original voice while fixing errors and adding structure. You’ll also learn why Grok often works better than ChatGPT for this specific task.

Time to complete: ~15 minutes per batch of notes

---

Working with Transcribed Content

At this stage, you have collected your raw material: thoughts exported in text format. Whether you typed directly into text files or transcribed audio notes, the important thing is that you now have a collection of these files.

In this step, we’ll focus on formatting these notes properly.

The process requires uploading your text files to an AI assistant.

For this purpose, I personally use Grok to save money instead of ChatGPT. While you can certainly use ChatGPT, I’ve found that Grok delivers better results in this specific context, as ChatGPT tends to hallucinate more frequently and sometimes changes phrases unnecessarily. Grok seems to work better with shorter prompts. These are my current observations, though models are constantly evolving, so this information might already be outdated.

I’m confident that with a carefully crafted longer prompt, you could train ChatGPT to perform equally well, but I find that Grok with a simple prompt instructing it not to paraphrase but only fix errors works effectively.

Prompt for Formatting Notes

In this chat, I will send text that was transcribed from audio
notes. Your task is to prepare a clean, error-free transcript
formatted for plain-text copying.

Do not paraphrase or reword the content — only correct words that
are clearly misplaced due to incorrect speech recognition.

Format the text for clarity and structure: use paragraph breaks,
numbering, or bullet points where applicable.

The names of the text files with transcripts include the date and
time of the note — arrange the final transcript in chronological
order accordingly.

Do not change the original thoughts — only fix obvious
transcription errors.

The Formatting Process

After uploading your text files to the AI assistant, it will produce a properly formatted version. The raw transcripts typically come as flat text without paragraph breaks, but AI can divide the content into paragraphs, add structure, create headings, and apply appropriate formatting. I prefer dividing the text into paragraphs (short semantic blocks), adding proper punctuation, correcting errors, removing duplicates, and fixing incorrectly recognized words.

The result is a well-formatted text divided into paragraphs, ready for use. I call this the “source” — the foundation upon which the next stages will be built.

In the same chat session, you can continue uploading new notes that you create while developing your ideas. The AI will simply continue formatting them accordingly.

This is Lesson #12 of the ANTIghostwriter course — a free, complete system for creating authentic content with AI assistance.

New here? Start from the full course overview.

Previous lesson: #11: My Morning Walk Method for Creating Raw Content

---

What You’ll Learn

Set up free, offline voice transcription using OpenAI’s Whisper model. This bonus lesson includes step-by-step installation instructions and a Python script that batch-processes your audio notes into text files automatically. Whisper works in 20+ languages with near-perfect accuracy — and it runs entirely on your computer, no internet required.

Time to complete: ~45 minutes for initial setup (technical)

---

As a bonus feature, I recommend using note transcription. Since I mentioned that I record text in audio format and then transcribe it, I suggest using the same system that I use.

For note transcription, you need AI and the Whisper model from OpenAI, which you can download locally for free and use on your own computer. It works offline, meaning it doesn’t require an internet connection, and it recognizes multiple languages and even accents.

Below I provide you with two resources:

1. A prompt that helped me set up an AI model on my computer. It will guide you step by step, and you’ll end up with a model that transcribes audio files into text.
2. A file containing a Python program that performs note transcription using the model installed with the help of the previous prompt.

For this system to work, you need to export your audio notes to a specific folder specified in the script and then run the script. I do this every morning: I record audio notes on my computer, export them to a folder, run the script, and it transcribes the audio and places text files in txt format next to the original audio files.

Prompt for Installing the Whisper Recognition Model from OpenAI

Help me install the OpenAI Whisper model on my computer. First,
guide me through gathering all the information we need to know as
a prerequisite to installation, then do the installation itself.

Do it step by step, one step at a time, so I can send you the
result after each step.

Transcription Script

This is a small program that performs transcription of notes located in a specific folder.

Important! This script is designed to work on the Windows operating system. If you have macOS, ask the AI to adapt the program to your operating system within the same chat.

import os

# Set personal cache directory (inside the user's home folder)
os.environ['XDG_CACHE_HOME'] = os.path.expanduser(r"~\.whisper_cache")

import whisper
from pathlib import Path

# Path to the folder containing your audio files
# Replace this with the actual path to your own audio files directory.
# This can be a local path (e.g., "C:/Users/yourname/audio") or a
# network share like the example below.
ROOT_DIR = Path(r"\\192.168.1.115\media\audio\notes")

# Supported audio file extensions
AUDIO_EXTENSIONS = {'.mp3', '.wav', '.m4a', '.ogg', '.webm', '.flac'}

# Load the Whisper "medium" model on GPU (if available)
# You can change "medium" to another model name like "tiny",
# "base", "small", or "large"
# depending on your desired balance between speed and accuracy.
model = whisper.load_model("medium", device="cuda")

# Recursively scan the folder for audio files that do not yet
# have transcripts
for audio_path in ROOT_DIR.rglob("*"):
    if audio_path.suffix.lower() not in AUDIO_EXTENSIONS:
        continue

    txt_path = audio_path.with_suffix(".txt")

    if txt_path.exists():
        print(f"✓ Already exists: {txt_path.name} – skipping")
        continue

    print(f"⏳ Transcribing: {audio_path.name}")

    try:
        # ✓ Automatically detects the language of the audio
        # ✓ If you know all files are in a specific language, you
        # can uncomment the next line and set the language manually
        # result = model.transcribe(str(audio_path), language="ru")
        result = model.transcribe(str(audio_path))  # automatic language detection

        text = result["text"]

        with open(txt_path, "w", encoding="utf-8") as f:
            f.write(text)

        print(f"✓ Saved: {txt_path.name}")

    except Exception as e:
        print(f"✗ Error with {audio_path.name}: {e}")

input("Press Enter to exit…")

This is Lesson #11 of the ANTIghostwriter course — a free, complete system for creating authentic content with AI assistance.

New here? Start from the full course overview.

Previous lesson: #10: Generate 50+ Content Ideas with AI in Minutes

---

What You’ll Learn

My personal method for creating raw content: the morning walk routine. Wake up, go for a walk, meditate for 15 minutes, then record voice notes for the rest. By the end of a typical walk, I have 3,000-4,000 words of authentic content. On longer weekend walks, up to 7,000. This lesson covers structuring your thoughts while speaking and why authentic raw material beats AI-generated content.

Time to complete: ~30 minutes to try the method once

---

Now that you have an idea for creating content, it’s time to dump your thoughts into notes. This is the method we discussed in lesson 8, where you use one of your tools to capture your thoughts.

You’ll start writing an article—I won’t dictate how to do this, as it’s your personal routine and method. Personally, I use the walking method. The first thing I do after waking up and using the bathroom is go for a walk.

My Personal Method

I begin with a morning meditation during the first 15 minutes of my walk, which helps me set the tone for the day and fully wake up. The light cardio from walking helps my body become alert. Additionally, I make a point to observe the sunrise, allowing me to get my first sunlight exposure of the day and wake up in harmony with nature.

During this time, my mind is clear—there’s no mental clutter yet, as I haven’t consumed any content. I can freely express my thoughts, which are typically already structured and well-formed. This is where my ideas emerge. This is my content.

I turn on a voice recorder and dictate my stream of thoughts. By the end of the walk, I’ve generated a substantial amount of material which I later transcribe. It usually amounts to several thousand words—3-4 thousand from a single walk. On weekends when I walk longer, I can produce up to 7 thousand words.

Why Authenticity Matters

This brings us back to why it’s important to use AI as an editor, not as a content creator. By expressing our own thoughts, we maintain authenticity. We’re not creating content from scratch or shifting responsibility to AI. We use AI to refine our thoughts, which is what we’ll explore next.

At this stage, you need to create original content, even if it’s rough. The language you use doesn’t matter. I create content in Russian because my Russian is more developed—I grew up with it and my vocabulary is broader. I can create beautiful speech patterns in Russian that I can’t achieve in English due to my more limited vocabulary and knowledge of the language.

In Russian, I’ve read great writers like Dostoevsky and Tolstoy, who mastered the art of beautiful writing. Language becomes richer when reading such books. That’s why I use Russian for writing my original articles.

Write in whatever way works best for you. If English works better, use English. It’s better to produce more content than less, because the foundation of good content is deep thinking. A deep thought can be compressed into a 280-character post, but a shallow thought cannot be expanded into a meaningful article—AI will simply invent content that may not align with your ideas.

Content Structure Framework

It’s beneficial to use your natural thought process and immediately structure it according to the framework you’re using. For example:

1. Present a problem that a person experiences. Describe the situation and establish the pain point (you can slightly emphasize for impact).
2. Briefly mention that a solution exists, making the reader aware.
3. Delve deeper into the topic: explain why you reached this conclusion and how it helped in real life.
4. Provide a step-by-step plan of what needs to be done to achieve similar results, or what you did.
5. Draw a conclusion.

This is a ready-made structure from which you can build posts and create content. The structure is refined into a polished format with the help of AI.

Final Advice

Create a substantial initial draft. Don’t worry if your speech is awkward, if you make mistakes, or if you repeat yourself. That’s precisely what AI is for—to eliminate repetitions, fix errors, and structure the text into a readable and easily digestible format, which we will cover in future lessons.

This is Lesson #10 of the ANTIghostwriter course — a free, complete system for creating authentic content with AI assistance.

New here? Start from the full course overview.

Previous lesson: #09: The 8-Step Content Creation Cycle Explained

---

What You’ll Learn

Never run out of content ideas again. In this lesson, you’ll learn to use your expertise and interests as endless idea sources, apply the “four eternal markets” framework (health, wealth, relationships, happiness) to make any topic resonate, learn from content you already consume, and use AI to generate 50+ ideas in minutes based on your customer avatar.

Time to complete: ~20 minutes to generate your first batch of ideas

---

Generating Content Ideas: How to Create a Never-Ending Library

Content idea generation is a common challenge: how do you create a huge amount of content to build a library that never ends? Everyone eventually finds their own method for generating ideas, so recommendation number 1 is to create your own system that works for you. Until you develop that system, you can use the following methods.

If you’ve chosen to pursue your interests for personal brand positioning, you shouldn’t face a deficit of ideas. Simply talk about what interests you. I chose a topic that’s broad enough—philosophy and reasoning—something I can discuss for hours every day. These naturally become my content ideas.

Your expertise and interests are the first source of ideas. The key is to use frameworks and structure your content to make it interesting to others.

How to Make Content Interesting

Wrap your content in motivators that resonate with people. I recommend reading my article about the four eternal markets: health, wealth, relationships, happiness, plus the fifth, spirituality. These represent fundamental needs that humans have and always will have in our current incarnation as a species.

Address pain points in each content element—this makes it interesting. Don’t just talk about a topic; frame it so that it impacts, for example, health. Or discuss how to approach relationships from a status perspective. The status game provides great positioning for content.

Learn from Others

Study the content you consume: what do you find interesting to watch or read? These are ideas you can borrow. Take an idea that a YouTuber discusses and filter it through your unique prism of knowledge, skills, and experience. This will create unique content. There may be some overlap, but the key is to capture the idea and tell it from your perspective. For someone who resonates with your voice and delivery style, your content will land better.

You might disagree with some perspectives. One person’s ideas resonate differently with another. Do your own thing—create, adapt ideas, and combine several concepts.

Your favorite books, podcasts, and all consumed content can be sources to extract ideas from. While reading a book, a single phrase or statistic can become the foundation for a video or article that you can elaborate on in detail and use as a content unit.

Generating with AI

1. Use Kortex, which has tools for generating ideas based on previously created customer avatars.
2. Upload your reader avatar to AI, for example, in ChatGPT, press search, ask it to find acute pain points of the target audience, popular ideas, videos, and suggest material for ideas.

These ideas can become headlines for your content units, posts, or scripts for videos.

AI is particularly helpful if you’re experiencing writer’s block and don’t know where to start. It easily overcomes this block. Make use of it.

At this stage, you only need one idea on which you’ll create articles and posts. Remember that this is a cyclical process. If you have several ideas at once, save them in your knowledge base and return to them in the next iteration.

This is Lesson #09 of the ANTIghostwriter course — a free, complete system for creating authentic content with AI assistance.

New here? Start from the full course overview.

Previous lesson: #08: Three Methods to Capture Your Thoughts for Content

---

What You’ll Learn

The 8-step cyclical process you’ll repeat for every piece of content: generate ideas → dump thoughts → format notes → research → write article → create posts → build threads → (optional) add images, SEO, scheduling. This visual overview shows how all the pieces fit together. Starting next lesson, we’ll dive deep into each step.

Time to complete: ~3 minutes to understand the cycle

---

The Content Creation Cycle

The following section outlines the core cyclical process that you’ll implement daily to build your content library. Starting with the next step, all elements will repeat in a continuous cycle:

1. Generating ideas
2. Offloading thoughts into notes
3. Formatting notes
4. Researching the idea
5. Creating articles
6. Creating social media posts
7. Developing content threads
8. Optional:
   • generating images for articles
   • adding SEO elements
   • planning future posts

Our approach is deliberately cyclical and repetitive. The diagram below illustrates the complete cycle, which we’ll explore step by step throughout this lesson.

This is Lesson #08 of the ANTIghostwriter course — a free, complete system for creating authentic content with AI assistance.

New here? Start from the full course overview.

Previous lesson: #07: Generate Your Writing Style Guide with Claude AI

---

What You’ll Learn

Three methods for getting ideas out of your head: handwriting (most authentic, most labor-intensive), typing (easiest to format), and audio notes (my personal favorite — I create 3,000-7,000 words during morning walks). Choose the method that fits your workflow. The goal is capturing your authentic thoughts as raw material for AI to refine.

Time to complete: ~5 minutes to read, ongoing to practice

---

Ways to Unload Your Thoughts

Methods for getting your thoughts out of your head and onto something tangible. This is a particularly important topic for me personally, because this is where you actually need to externalize your thoughts. You’ll be creating content on various topics. We’ll discuss idea generation for these topics in later lessons. Here, I’m focusing on the actual format and methods for writing content.

1. Handwriting

Some people write with a pen or pencil in a notebook or on paper. For many, this is the most effective method. It’s authentic because it engages multiple senses: fine motor skills, muscle memory, visual feedback (seeing the text as you write), auditory input (hearing the pen scratch against paper), and even smell (the scent of ink). This multi-sensory engagement allows for deeper immersion in the writing process. In terms of physical sensations, it’s the most authentic way to capture thoughts. Many writers prefer this method, but it’s labor-intensive, time-consuming, and doesn’t allow for easy automation or digital processing.

2. Typing Tools

This includes keyboards and digital document creation tools. You might use a typewriter, laptop, computer, or software like Word, Notion, Kortex, Obsidian, or other applications where you record your thoughts digitally. It’s often better to capture thoughts this way immediately because it’s easy to format text, create lists, add headings, and highlight important words. This method is highly convenient for organization and editing.

3. Audio Notes

I create most of my materials using audio notes. This is an accessible option that can be combined with other activities like walking. I often develop content during walks, when my brain works at its best. I walk and dictate text into a voice recorder, then later convert it to text and format it into articles. This isn’t the simplest process, as audio notes need to be transcribed and formatted. My workflow involves recording thoughts in audio, transferring them to my computer, using an AI model to transcribe the speech into text, then using additional AI models to format the content, break it into paragraphs, and correct errors. Raw speech often contains repetitions and mistakes—AI helps remove these distracting elements.

---

Choose a method that feels comfortable and suits your workflow. The goal is to produce content that will serve as a foundation for posts, articles, scripts, and other materials.

This is Lesson #07 of the ANTIghostwriter course — a free, complete system for creating authentic content with AI assistance.

New here? Start from the full course overview.

Previous lesson: #06: Create Your Customer Avatar with AI Interview Process

---

What You’ll Learn

The most important document in your content system: a style guide that lets AI write in YOUR voice. In this lesson, you’ll use a comprehensive Voice Analysis Prompt to analyze your existing writing and produce a detailed guide covering your vocabulary, sentence structure, tone, persuasion techniques, and more. This document becomes the foundation for all AI-assisted content.

Time to complete: ~60 minutes to gather samples and run analysis

---

This lesson covers how to create a style guide that allows anyone to write in your unique voice. Your personal style includes characteristic phrases, speech patterns, and vocabulary that make your writing distinctively yours.

To create such a guide, you first need writing samples that demonstrate your voice. Then you’ll analyze these samples using a specialized prompt that identifies the patterns and elements that define your writing style.

How It Works

The process takes content written by you—short posts, long texts, or even transcribed voice notes—primarily in text format. You feed these samples to an AI tool using the voice analysis prompt described below. The AI will analyze your writing style and produce a comprehensive guide that can be used for creating future content that matches your voice.

Voice Analysis Prompt

Purpose

For this exercise, create an EXTENSIVE writing style guide so
that ANYBODY can write the exact same way as the author of the
reference writing without ever reading any of their writing
directly. This guide should be built incrementally with each new
sample provided.

Important Instructions

— If I do not provide a reference sample upfront, ask me to
provide one.

— Each post I give you should be ADDED INTO the previous
analysis, do not REPLACE, or REMOVE A SINGLE WORD of the previous
examples.

— Extract patterns without losing the distinctive qualities that
make this writing unique

— Provide concrete techniques that can be learned and applied

— Reference samples are in Russian, but the analysis must be in
English

Analysis Categories

1. Lexical Choices

Compile a comprehensive lexicon of frequent words and phrases
used, emphasizing:

— Distinctive vocabulary choices
— Characteristic verbs and how they drive action
— Adjectives and adverbs that create an emotional tone
— Recurring phrases or expressions that define their voice
— Specialized terminology or jargon (if applicable)
— Word length preferences and vocabulary complexity level

(Include verbatim examples that best reflect their writing for
each lexical pattern identified)

2. Sentence Structure

Analyze and categorize typical sentence constructions, including:

— Average sentence length and variation patterns
— Use of fragments vs. complete sentences
— Comma usage and conjunction patterns
— Question or exclamation frequency
— Use of parenthetical asides
— Short, punchy statements for impact versus longer, flowing
sentences for storytelling
— Passive vs. active voice preferences

(Include verbatim examples that best reflect each structural
pattern in their writing)

3. Opening Lines

Document and dissect the most effective opening lines to
understand engagement techniques:

— Question vs. statement vs. exclamation opening strategies
— Length of typical opening sentences
— How they introduce the subject matter
— Provocative vs. explanatory approaches
— Direct address vs. third-person narration
— How they establish authority or connection immediately

(Include verbatim examples of opening lines, with analysis of
what makes each effective)

4. Flow and Rhythm

Break down the rhythmic qualities of paragraphs and transitions:

— Paragraph length patterns
— Transition techniques between paragraphs
— Sentence length variation patterns within paragraphs
— Use of one-sentence paragraphs and their strategic placement
— Pacing changes throughout pieces (acceleration/deceleration)
— Cadence and how it creates emphasis

(Include verbatim paragraph examples showing flow variations and
rhythmic techniques)

5. Persuasive Techniques

Outline specific persuasive strategies employed:

— Use of ethos, pathos, and logos
— Rhetorical questions and their placement
— Direct reader address methods
— Call-to-action structures
— Social proof incorporation
— Repetition and emphasis patterns
— Story frameworks and narrative techniques
— Problem-solution structures
— Use of statistics and external references

(Include verbatim examples demonstrating each persuasive
technique)

6. Formatting Styles

Document distinctive formatting preferences:

— Paragraph structure and length patterns
— Use of white space and its strategic purpose
— Text emphasis techniques (bold, italics, CAPS, etc.)
— List structures (bulleted, numbered, checked)
— Section headers and subheadings approach
— Quote formatting and attribution
— Link placement and anchor text patterns
— Use of special characters or symbols

(Include verbatim examples of formatting choices with notes on
their effectiveness)

7. Emotional Appeal

Analyze emotional connection strategies:

— Primary emotions targeted (curiosity, fear, excitement, etc.)
— How tension and resolution are balanced
— Use of contrast between emotional states
— Personal disclosure patterns
— Reader validation techniques
— Creation of exclusivity or insider knowledge
— Vulnerability vs. authority balance
— Humor usage patterns and types

(Include verbatim examples showing emotional appeal techniques)

8. Tone and Voice

Document the distinctive tone characteristics:

— Formal vs. conversational balance
— Humor type and frequency
— Sarcasm or irony usage
— Authoritative vs. collaborative stance
— First/second/third person usage patterns
— Consistency vs. variation across different topics
— Cultural references and their function

(Include verbatim examples demonstrating tone and voice
qualities)

9. Miscellaneous Distinctive Elements

Identify any other unique writing characteristics:

— Signature phrases or linguistic "tics"
— Metaphor and simile patterns
— Storytelling frameworks
— Introduction of characters or case studies
— Meta-commentary on the writing itself
— Breaking of conventional writing rules

(Provide core elements of their writing verbatim, and annotate
explaining why certain choices were made and how they contribute
to the overall effectiveness)

Synthesis

After compiling examples in all categories, create:

1. A one-page "quick reference guide" with the most essential
elements

2. A step-by-step process for emulating this writing style

3. A before/after example showing how a generic paragraph would
be transformed using this style

Application

For maximum learning impact, include recommendations for:

— Practice exercises to develop mastery of this style

— Common pitfalls to avoid when emulating this writer

— How to adapt the style for different contexts while maintaining
authenticity

Getting Started

To begin this exercise, you’ll need to create or gather content samples. Here are some approaches:

• If you’ve made previous attempts to establish your brand, collect your existing articles or posts from social media platforms. Choose those that best reflect your authorial style.
• If you don’t have suitable existing content, write a few articles or posts that express your thoughts in your natural voice. Pick a topic you’re knowledgeable about and describe it in your own words.
• Another approach is to record yourself speaking about a topic you’re passionate about. Transcribe this recording and use the text as material for the analysis.

You can use these initial samples to generate your voice analysis, which becomes your starting point. Later, you can add more content to the analysis to further refine your authorial style guide.

An important consideration: you might want to incorporate elements of someone else’s style. For example, if you admire how a particular person writes or expresses their thoughts, you can include their material in your analysis. The AI will incorporate these elements into your style guide.

Personally, I prefer developing my authentic voice as it’s part of my personal brand, but you’re free to approach this however you wish.

Saving Your Style Guide

The end result of this exercise should be a document containing your authorial style guide. Save this document just as you did with the reader avatar.

These elements can be saved as separate documents—as pages in Notion or Kortex, Word files, or whatever format works for you. It’s best to convert them to PDF format for future use. Kortex and Notion allow you to do this easily, or you can convert Word documents to PDF. If you’re working with Claude, which I recommend for creating both the avatar and authorial style, there’s also a PDF export function.

By the end of this lesson, using the provided prompt, you should have a comprehensive guide to your authorial style.