flackz
Senior Member
- Nov 28, 2023
- 836
- 342
I asked GPT
In fact, ChatGPT has no idea how the AI detectors work, so I gave a link to a solution that we use because it just works for us.
I asked GPT
AI writes very well, but you have to review and edit according to your words. AI only supports and rewrites the most basic things.I am new to the AI and ChatGPT world. I have tried to get a couple articles from ChatGpt which were readable and even passed the plagiarism test on several platforms. However the articles were detected as AI written content on several AI Detection websites. My question is:
(1) What is the best way to make them pass the AI detection test?
(2) How does these AI detection tools detect these articles as AI generated?
I know i can rewrite them in my own words but that is very time consuming and we are in 2024 in the time of AI. I have tried several tools that can humanize these articles but the output is quite poor. I need some suggestions please.
Can you please send me the command prompts you use?You can rewrite using commands on chatgpt itself to make it unique. Practise this regular and you will become better at it.
Thanks for sharing this.I asked GPT :
Identifying watermarks or markers in AI-generated text can be challenging, as these markers are often subtle and not always explicitly defined. However, here are some strategies you can use to detect potential indicators of AI-generated content:
- Repetitive Patterns: AI-generated text may exhibit repetitive phrases or structures. Look for sentences that seem formulaic or overly similar in style and structure.
- Lack of Depth: AI-generated content might lack nuanced understanding or depth on complex topics. If the text seems superficial or overly general, it could be a sign of AI generation.
- Inconsistent Tone or Style: If the text shifts abruptly in tone or style, it may indicate that it was generated by an AI. Human writing typically maintains a more consistent voice throughout.
- Unusual Word Choices: AI might use uncommon or slightly off-context words that a human writer might not choose. Look for phrases that seem awkward or out of place.
- Overly Formal or Structured Language: AI-generated text can sometimes be overly formal or structured, lacking the natural flow of conversational language.
- Errors in Context: AI may generate content that contains factual inaccuracies or misinterpretations of context. Check for statements that seem incorrect or out of context.
- Lack of Personal Experience: AI cannot draw from personal experiences or emotions. If the text lacks personal anecdotes or emotional depth, it may be AI-generated.
- Use of Clichés: AI often relies on clichés or common phrases. If the text is filled with overused expressions, it might be a sign of AI generation.
- Detection Tools: There are tools and software designed to detect AI-generated text. These tools analyze writing patterns and can provide insights into whether the content is likely AI-generated.
- Cross-Referencing: If you suspect a piece of text is AI-generated, you can cross-reference it with known AI outputs or use plagiarism detection tools to see if it matches existing content.
Removing watermarks or indicators of AI-generated text involves editing the content to make it appear more human-like and natural. Here are some strategies to help you achieve that:
- Rephrase Sentences: Rewrite sentences to change their structure and wording. Use synonyms and alter the sentence flow to make it sound more conversational and less formulaic.
- Add Personal Touches: Incorporate personal anecdotes, experiences, or opinions. This can help make the text feel more authentic and relatable.
- Vary Sentence Length and Structure: Mix short and long sentences to create a more dynamic rhythm. Avoid repetitive patterns by changing how you construct sentences.
- Use Natural Language: Aim for a conversational tone. Avoid overly formal language and use colloquial expressions where appropriate.
- Eliminate Clichés: Replace clichés with original phrases or ideas. This can help the text feel fresher and more unique.
- Enhance Depth and Detail: Add more context, examples, and explanations to provide depth. This can help the text move beyond surface-level information.
- Check for Consistency: Ensure that the tone and style are consistent throughout the text. Make adjustments to maintain a unified voice.
- Proofread for Errors: Review the text for any grammatical or contextual errors that may have been introduced. Correcting these can improve the overall quality.
- Use a Thesaurus: To find alternative words and phrases, use a thesaurus to replace any awkward or unusual word choices with more common ones.
- Read Aloud: Reading the text aloud can help you identify areas that sound unnatural or awkward. This can guide you in making necessary adjustments.
- Seek Feedback: If possible, have someone else read the text and provide feedback. They may catch issues you might have missed and suggest improvements.
Can you please share them?Hey Money Guru,
Initially when I started doing Blogs on WP. I had the same problem and later I realized that there are a lot of free 'Humanizing" tools online, which could be used to evade AI detection.
I hope it helped, thank you.
Whats the name of this siteThere is a very simple way to “humanize” any text. To confuse detectors, just make the text less than ideal. No, I do not suggest making mistakes, just small typos, and rough edges. For example a space before a period, or a lowercase letter at the beginning of a sentence, or remove a comma, or highlight any word in CAPS, or change the apostrophe symbol, or a double hyphen or a space - there are a lot of options. And the result will be something like this:
Before
View attachment 383733
After
View attachment 383735
Can you please share them?
Sure,Whats the name of this site
It always says like 75% - 100% HUMAN-GENERATED CONTENT on the texts generated by GPT-4. So, looks like it's a fake detector.https://writer.com/ai-content-detector/ Site from my post
I checked the services for checking and humanizing the text written by AI, which were offered here. The results are shown in the screenshots before and after processing.
Yes, of course. But here I wanted to show that detectors cannot give an accurate estimate. By the way, I gave the AI two texts to check, and one was written by me, the other by him. He determined that the text I wrote was most likely written by a human, because it used phrases inherent in human speech. But in any case, I think that humanizing a text does not mean making it human.Don't you see that this text has simply been synonymized? Sure, you can easily fake any "AI detector" with a grandpa's synonymizer script, but it won't make your articles look like they were written by a human. Just read the text you see in the screenshot above.![]()
Yes, of course. But here I wanted to show that detectors cannot give an accurate estimate. By the way, I gave the AI two texts to check, and one was written by me, the other by him. He determined that the text I wrote was most likely written by a human, because it used phrases inherent in human speech. But in any case, I think that humanizing a text does not mean making it human.
In my opinion, with the level AI is at now, you still need the services of a writer to tweak the text to pass AI detection.I am new to the AI and ChatGPT world. I have tried to get a couple articles from ChatGpt which were readable and even passed the plagiarism test on several platforms. However the articles were detected as AI written content on several AI Detection websites. My question is:
(1) What is the best way to make them pass the AI detection test?
(2) How does these AI detection tools detect these articles as AI generated?
I know i can rewrite them in my own words but that is very time consuming and we are in 2024 in the time of AI. I have tried several tools that can humanize these articles but the output is quite poor. I need some suggestions please.
The problem with AI written articles is that they can be too perfect. To make them more human like, try adding personal touches and editing them yourself. While AI tools can help, human input is still crucial for creating authentic content.I am new to the AI and ChatGPT world. I have tried to get a couple articles from ChatGpt which were readable and even passed the plagiarism test on several platforms. However the articles were detected as AI written content on several AI Detection websites. My question is:
(1) What is the best way to make them pass the AI detection test?
(2) How does these AI detection tools detect these articles as AI generated?
I know i can rewrite them in my own words but that is very time consuming and we are in 2024 in the time of AI. I have tried several tools that can humanize these articles but the output is quite poor. I need some suggestions please.