How to Create Human-Like Text as a Real Writer, and Leave AI Behind

How to Stand Out as a Writer Against AI and actual Learn to be Interesting in your writing, and write fast, interesting content, that ranks better and actually helps people

Journalists against generic content have banded together to create the official guideline of the human writer.

  1. Do Not Rely Solely on AI for Entire Articles, nor even a little bit.

    • AI tools are best used for generating small sections of text or overcoming writer’s block. Creating entire articles with AI can result in repetitive and incoherent content that is easy to detect.
  2. Structure and Format Properly
    • Break up long paragraphs, use sufficient white spaces, and include images or emojis where appropriate. This reduces the likelihood of detection by making the content appear more human-like.
  3. Diversify Vocabulary
    • Avoid repetitive words and phrases. Use synonyms and more conversational language, including contractions. This makes the text less formulaic and more varied.
  4. Paraphrase and Edit
    • Rewriting AI-generated content manually or using paraphrasing tools can help in making it sound different while keeping the core idea intact. Adding transitional sentences can improve the flow and readability.
  5. Match the Tone and Style
    • Ensure the content matches the intended tone and style of the brand or author. Using prompts like “write in the style of [author/blogger]” can help in generating more personalized outputs.
  6. Include Human Elements
    • Incorporate personal anecdotes, idiomatic expressions, and emotional language. This adds a human touch to the content.
  7. Intentional Variations and Inconsistencies
    • Introducing minor spelling variations, using both US and UK English spellings, and varying sentence lengths can help in making the text appear more human-like.

Key Findings on AI vs. Human Writing

  1. Complexity and Length
    • Human-written paragraphs tend to be longer and more complex. Humans also vary their sentence structures more than AI, using a mix of short and long sentences.
  2. Punctuation and Word Usage
    • Humans use more punctuation marks like semicolons, dashes, and question marks. They also use more proper nouns, numbers, and specific names, whereas AI prefers general terms like “researchers.”
  3. Equivocal Language
    • Words like “however,” “but,” and “although” are more frequently used by humans. AI-generated text tends to avoid such nuanced language.
  4. Stylistic Differences
    • AI-generated content often lacks the stylistic diversity seen in human writing. This includes less variation in sentence length and structure.

Practical Implications

  • Google’s Stance
    • Google has stated that it does not penalize AI-generated content as long as it is high-quality and provides value to users. The focus should be on content quality rather than the method of creation.
  • Detection Techniques
    • Simple machine learning models can achieve high accuracy in distinguishing AI-generated text from human writing by focusing on features like paragraph complexity, sentence length diversity, and specific word usage.

Real-World Application

For instance, the study by Heather Desaire and colleagues (2023) demonstrated over 99% accuracy in distinguishing AI-generated academic science writing from human-generated content using a modest training set and off-the-shelf tools. This approach can be adapted and applied to various types of writing to ensure the authenticity of human authorship.

By following these guidelines and understanding the key differences between AI and human writing, content creators can effectively create human-like text that is less likely to be flagged by AI content detectors.