Chat GPT - Pass Detection 100% Human Written With This Prompt

Success With AI
30 Jan 202306:08

TLDRIn this video, the presenter shares a method to bypass AI content detectors without additional software. Initially, an article written by Chat GPT was detected with 97% confidence as AI-generated by Originality.ai and 90% human-generated by Writer.com. To improve the article's human-likeness, the presenter instructs Chat GPT to write with increased perplexity and burstiness, which are key factors in assessing content's AI nature. After rewriting the article with these instructions, it was re-evaluated and received a 100% human content score on all three AI detectors. This technique offers the advantage of not needing extra tools and is less likely to degrade the quality of well-written content. The presenter encourages viewers to test this method and share their results.

Takeaways

  • 🔍 **Language Identification:** The language of the title is English.
  • 📝 **Content Detection Tools:** The video discusses using AI content detectors like Originality.ai and writer.com to check for AI-generated content.
  • 🚫 **Avoiding Detection:** The presenter shares a method to bypass AI content detectors without additional software.
  • 💡 **Quill Bot Mention:** It's noted that Quill Bot, a paid tool, can help re-draft content to pass AI detection.
  • 📈 **Perplexity and Burstiness:** Two crucial factors in assessing AI content are perplexity (text complexity) and burstiness (variation in sentence length).
  • 💻 **Rewriting for Detection:** The video demonstrates how to use prompts to instruct Chat GPT to write with high perplexity and burstiness to avoid detection.
  • 🔄 **Rechecking Process:** After rewriting the content, it's rechecked with AI detectors to confirm its human-like nature.
  • 🆕 **Originality Score:** The rewritten article achieves a 100% originality score on AI detectors.
  • 📉 **No Additional Tools Needed:** The method allows for content rewriting without the need for extra tools.
  • 📚 **Maintaining Quality:** The approach is said to be less likely to degrade the quality of well-written content.
  • 📊 **Varying Results:** The success of the method may vary depending on the type of content, with results ranging from 80% to 100% human-like scores.
  • 🎥 **Viewer Engagement:** The presenter encourages viewers to like, comment, and subscribe for more information on achieving AI success.

Q & A

  • What is the main purpose of the video?

    -The main purpose of the video is to demonstrate how to bypass AI content detectors without using additional software.

  • What tool was mentioned for re-drafting content to pass AI detection?

    -Quill Bot is mentioned as a paid tool that can help re-draft content to pass AI detection.

  • What are the two factors crucial for rating content in the context of AI detection?

    -The two factors crucial for rating content are perplexity, which measures the complexity of the text, and burstiness, which compares the variations of the sentences.

  • What is the difference between human and AI written content in terms of burstiness?

    -Humans tend to write with greater burstiness, including longer or more complex sentences alongside shorter ones, whereas AI sentences tend to be more uniform.

  • What was the initial AI detection confidence level for the 250 word article written by chat GPT?

    -The initial AI detection confidence level for the 250 word article was 97%, indicating it was highly detectable as AI-generated content.

  • How did the video demonstrate modifying content to reduce AI detection?

    -The video demonstrated modifying content by adding a prompt to increase perplexity and burstiness, then rechecking the content with AI detectors.

  • What was the result of the rewritten article after increasing perplexity and burstiness?

    -The rewritten article scored 100% on all three AI detectors as human-generated content.

  • Which website was used to scan the article for AI detection?

    -The websites used to scan the article for AI detection were originality.ai and contentatscale.ai.

  • Why is it important to balance perplexity and burstiness when writing content to avoid AI detection?

    -Balancing perplexity and burstiness makes the content more complex and varied, mimicking human writing patterns, which is less likely to be detected by AI as machine-generated.

  • What is the advantage of not using an additional tool for content re-drafting?

    -The advantage is that it simplifies the process and potentially reduces costs, as well as maintaining the quality of well-written content without making it appear 'cruddy'.

  • How many times has the presenter tested the method of increasing perplexity and burstiness?

    -The presenter has tested the method at least a dozen times.

  • What is the recommendation for viewers interested in the video's content?

    -The recommendation is to like, comment, subscribe to the channel for more information on achieving AI success and to share feedback on how the method is working.

Outlines

00:00

🔍 Bypassing AI Detectors with Perplexity and Burstiness

The first paragraph of the video script introduces the viewer to a method for evading AI content detection without the need for additional software. The host discusses their previous content about using Quill Bot, a paid tool for re-drafting content to pass AI detection. They then present a new approach using Chat GPT to avoid detection. The host demonstrates this by writing a 250-word article on how Chat GPT can revolutionize the real estate industry and scanning it with Originality.ai, which detected it with 97% confidence as AI-generated. They then use Writer.com's AI detector, which surprisingly identified it as 90% human-generated. To improve the article's disguise against AI detection, the host instructs Chat GPT to rewrite the article with increased perplexity and burstiness, which are crucial factors in content assessment. After rewriting, the article is re-scanned and successfully identified as 100% original or human-generated by multiple AI detectors.

05:03

📈 Benefits of Enhanced Burstiness and Perplexity

The second paragraph emphasizes the advantages of the method just demonstrated. The primary benefit is the elimination of the need for additional tools, and the secondary benefit is that increasing burstiness and perplexity makes the content less likely to be detected by AI as machine-generated. The host shares their testing experience, noting that the method has been consistently effective across different trials, although the success rate can vary depending on the content type. The paragraph concludes with an encouragement for viewers to try the method, engage with the channel through likes and comments, and to subscribe for more information on achieving success with AI.

Mindmap

Keywords

💡AI content detectors

AI content detectors are algorithms or software programs designed to identify content that has been generated by artificial intelligence rather than by human authors. In the video, the focus is on bypassing these detectors to make AI-generated content appear as if it were written by a human. This is important for content creators who wish to avoid potential penalties or restrictions that may be associated with AI-generated content.

💡Quill bot

Quill bot is a paid tool mentioned in the video that assists in re-drafting content to help it pass AI detection. It is an example of a service that can be used to make AI-written content seem more human-like, thus reducing the likelihood of it being flagged by AI content detectors.

💡Perplexity

Perplexity, in the context of the video, refers to the complexity of the text. It is one of the two factors, along with burstiness, that AI content detectors use to assess whether content is likely to be AI-generated. High perplexity indicates a more complex and varied structure in the text, which is characteristic of human writing.

💡Burstiness

Burstiness is the second factor used by AI content detectors to evaluate content. It compares the variations in sentence length and complexity. Human writing tends to have greater burstiness, with a mix of longer, more complex sentences and shorter, simpler ones. In contrast, AI-generated sentences are often more uniform and less varied.

💡Originality.ai

Originality.ai is an AI detection tool mentioned in the video. It assesses the likelihood that a piece of content is AI-generated by analyzing its originality and complexity. The video demonstrates how the confidence score of AI detection can vary between different tools, with Originality.ai initially showing high confidence that the content was AI-generated.

💡Writer.com

Writer.com is another tool referenced in the video that features an AI detector. It analyzes content to determine the percentage of human-generated content. Interestingly, it rated the AI-generated content as 90 percent human-generated after certain adjustments were made, suggesting a lower aggressiveness in detection compared to other tools.

💡Content at Scale

Content at Scale is a free AI detector that the video uses to test the rewritten article. It provided a 100 percent confidence score that the rewritten content was AI-based before adjustments were made. After increasing perplexity and burstiness, the tool's confidence in the content being AI-generated dropped significantly.

💡Rewriting with perplexity and burstiness

The process of rewriting content with increased levels of perplexity and burstiness is a technique demonstrated in the video to help AI-generated content evade detection by AI content detectors. By incorporating more complex sentence structures and varying sentence lengths, the rewritten content appears more human-like.

💡Human-generated content

Human-generated content refers to material that is created by human authors as opposed to AI. In the context of the video, the goal is to make AI-generated content appear as human-generated to fool AI detectors. The video shows that by adjusting for perplexity and burstiness, the content can score as 100 percent human-generated on certain detectors.

💡Avoiding detection

Avoiding detection is the main theme of the video. It discusses methods to make AI-generated content pass undetected by AI content detectors. This is done by adjusting the writing style to mimic human writing patterns, specifically by increasing perplexity and burstiness in the text.

💡AI success

AI success, as mentioned in the video, implies the effective use of AI tools and techniques to achieve desired outcomes, such as generating content that can evade AI detectors. The video provides a method to achieve this by manipulating the characteristics of AI-generated text to make it appear more human-like.

Highlights

The video demonstrates how to bypass AI content detectors without additional software.

Previous videos discussed using Quill Bot, a paid tool for re-drafting content to pass AI detection.

A 250-word article was written using Chat GPT on how it could revolutionize the real estate industry.

Originality.ai was used for AI detection, with a 97% confidence that the content was AI-generated.

Writer.com's AI detector claimed the content was 90% human-generated, indicating less aggressive detection.

Content at Scale AI detector showed 100% confidence that the content was AI-based.

The video explains the importance of using prompts and seed content to inform Chat GPT before writing.

Two crucial factors in content assessment are perplexity and burstiness.

Perplexity measures text complexity, while burstiness compares sentence variations.

AI sentences tend to be more uniform, whereas human sentences show greater burstiness.

The video shows how to rewrite an article with increased perplexity and burstiness to avoid AI detection.

Rewritten content using the new approach scored 100% as original on Originality.ai.

Writer.com's analysis also rated the rewritten content as 100% human-generated.

The free AI detector at Content at Scale rated the content as 100% human content.

The method does not require additional tools and maintains the quality of well-written content.

Increasing perplexity and burstiness makes content less likely to be detected as AI-generated.

The presenter has tested this method a dozen times with similar successful results.

The effectiveness may vary depending on the type of content.

Adding extra prompt changes sentence structure variations, helping to evade AI detectors.

The video encourages viewers to like, comment, subscribe, and share their experiences with the method.