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

Подписаться 52 тыс.

Просмотров 785 тыс.

50% 1

💥 Replicate ANY site with AI: Use this code to start with 10% off: SuccessWithAI10
✅ rebrand.ly/web10
🔥 Skyrocket Organic Traffic With Surfer SEO
👉 rebrand.ly/gosurferai
🔥 All In One Funnel Building Software:
👉 rebrand.ly/systemeio-free
🔥 Best Chat GPT SEO Content Writer Feb 2023:
👉 rebrand.ly/zimmwriter
Chat GPT - Pass Detection 100 Human Written With This Prompt
PROMPT SHOWN IN VIDEO:
"When it comes to writing content, two factors are crucial, "perplexity" and "burstiness." Perplexity measures the complexity of text. Separately, burstiness compares the variations of sentences. Humans tend to write with greater burstiness, for example, with some longer or complex sentences alongside shorter ones. AI sentences tend to be more uniform. Therefore, when writing the following content I am going to ask you to create, I need it to have a good amount of perplexity and burstiness. Do you understand?"
🔥 Create super fast AI videos using Pictory:
👉 rebrand.ly/gopictory
🔥 Skyrocket Organic Traffic With Surfer SEO
👉 rebrand.ly/gosurferai
🔥 Software To Allow Text To Pass AI Detection: (Free Trial)
👉 rebrand.ly/pass-ai
🔥 Quillbot Software Shown In Video To Pass AI:
👉 quillbot.com/
🔥 Free Limited AI Detector Shown In Video
👉 contentatscale.ai/ai-content-...
🔥 All In One Funnel Building Software:
👉 rebrand.ly/systemeio-free
#chatgpt
#artificialintelligence
#affiliatemarketing
In this tutorial, learn how to use specific prompts to get AI tools like ChatGPT to generate 100% human-like content. Discover the secrets to creating natural language and fooling even the most advanced AI algorithms. Get ready to take your chatbot game to the next level! Watch now.
Affiliate disclaimer:
Keep in mind that we may receive commissions when you click our links and make purchases. However, this does not impact my reviews and product how-tos.

Хобби

Опубликовано:

29 янв 2023

Ссылка:

Скачать:

Готовим ссылку...

Добавить в:

Мой плейлист

Посмотреть позже

Комментарии : 1,6 тыс.

@UseAI Год назад

💰 Automated AI Affiliate Product Review Sites: 👉 rebrand.ly/aiwise 🔥 Best Chat GPT SEO Content Writer Feb 2023: 👉 rebrand.ly/zimmwriter

@jesusonaunicorn Год назад

translating this to other langueges dos not work sadly... Cuando se trata de escribir contenido, dos factores son cruciales: "perplejidad" y "explosividad". La perplejidad mide la complejidad del texto. Separadamente, la explosividad compara las variaciones de las oraciones. Los humanos tienden a escribir con mayor explosividad, por ejemplo, con algunas oraciones más largas o complejas junto con otras más cortas. Las oraciones de la IA tienden a ser más uniformes. Por lo tanto, al escribir el siguiente contenido que voy a pedirte que crees, necesito que tenga una buena cantidad de perplejidad y explosividad. ¿si entienes contesta con "..." this does not work same in german

@GabeDev Год назад

imagine just being someone who naturally writes with an AI-type of punctuation and grammar and getting expelled from university because they don't believe you wrote your essays

@UseAI Год назад

Imagine that!

@ebeneZr Год назад

At that point you might as well add the most random words and misspellings just to prove you wrote it

@eswarjuri Год назад

@@ebeneZr Wait until someone codes an A.I. that makes exactly that…

@Weaksoft Год назад

@@eswarjuri Don't even need AI to do that. Any decent programmer should be able to write a script that adds random typos and whatnot to a text input. There are reddit bots that do this type of thing already.

@matgaw123 Год назад

@@eswarjuri or replace some liters that are almost identical for human eye but. For ai for example lI

@larsjonasson2959 Год назад

If you are a student who knows more than one language just generate the text in the other language and manually translate it. You will then learn the subject, and the teachers will not be suspicious since you are using your own words and the text will not contain the bizarre artifacts that often occurs in AI texts.

@astronemir Год назад

The AI is best in English. So if hour essay is Swedish etc., it’s fine. But if you have to submit in English I would still generate in English.

@biggooba6706 Год назад

that makes no sense

@SebastianTalbierz Год назад

its quite bad in polish too. it doesn't understand how to use 'old polish' words what are ingrained deeply into the language and it seems odd without them.

@cinemamovies1916 Год назад

@@biggooba6706 what does make any sense He said if you generate a prompt in a different language and translate it, then it won’t get flagged as “written by ai”

@b-lotus5145 Год назад

@@astronemir It seems to be good in Spanish too

@Rafiki041087 Год назад

This is an impressive example, not just because of the main objective to have ChatGPT successfully generate more humanlike output, but it's also demonstrating how to use prompt engineering. Great work.

@UseAI Год назад

Thank you!

@colamity_5000 Год назад

"prompt engineering" fucking kill me dude

@btm1 Год назад

@@colamity_5000 not a lot of engineering going on...

@fox971000 Год назад

What is prompt engineering?

@colamity_5000 Год назад

@@fox971000 Its just optimizing the prompts you give to an AI so that it gives you results that are closer to what you want. In the context of ChatGPT that means "engineering" questions that the AI has a high likelihood of accurately interpreting. It's like the new version of effective google searching.

@racool911 Год назад

Lmao I love how the solution is basically telling the AI to write something that isn't so AI-like

@UseAI Год назад

Exactly LOL

@goncalocarneiro3043 Год назад

"You're too professional, be less professional."

@pc_screen5478 Год назад

Detecting AI content, especially text, is an uphill battle. It's already impossible to detect AI usage on short sentences (e.g "have a good day") since nothing gives away whether it could or couldn't be AI, and with longer sentences there will always be a prompt that causes the AI to write differently from how it usually writes. These detectors will need more and more text to be able to detect AI usage at which point they'll largely become obsolete

@endermaster08 Год назад

AND IT WORKS

@aenetanthony Год назад

@@pc_screen5478Yeah, I think the false positive rates on these things are so high not only because they’re fairly new but also because it’s simply impossible not to get lots of false positives. There might be 20 ways to conveys the same information in a sentence, but if you have enough randomness and the AI is well trained enough, it could probably give you 16 of those, whereas if you asked humans you would probably have to ask a lot of people to get the same amount. Point is, there will always be cases where an AI and a human say the same thing, and it will end up with either a false positive or a false negative.

@DevonWayne Год назад

For those lazy, this is the prompt he used: When it comes to writing content, two factors are crucial, "perplexity" and "burstiness." Perplexity measures the complexity of text. Separately, burstiness compares the variations of sentences. Humans tend to write with great burstiness, for example, with some longer or complex sentences alongside shorter sentences. AI sentences tend to be more uniform. Therefore, when writing the following content I am going to ask you to create, I need it to have a good amount of perplexity and burstiness.

@rynhs Год назад

@@jxke5457 also in the description

@ozhinz Год назад

its in the description dawg

@ziwuri Год назад

might wanna change "a good amount" to "a high amount" as well

@Ketchupman11 Год назад

My hero

@Vicious1337 Год назад

Bruh

@russellthorburn9297 Год назад

The video expertly showcases the use of "burstiness" and "perplexity" to prevent AI-generated text from being detected as such. The demonstration effectively highlights the cutting-edge techniques used in language processing. - ChatGPT :-)

@UseAI Год назад

🤣

@ziwuri Год назад

I could honestly immediately tell this was written by ChatGPT from the lack of, y'know, burstiness and perplexity.

@mellomojared Год назад

Russell, I admit. Your comment was written with AI

@PhelanVanFloof Год назад

@@ziwuri It does indeed look very AI-generated, but not because of the burstiness or perplexity. You'd need to have at least a few paragraphs to be able to measure those.

@solanaceous Год назад

@@ziwuri plot twist, your comment was written by ai 😳

@cjfitguy Год назад

*Keeping this here for myself (and everyone else):* 1. I'm going to give you some information before asking you to write an article. Do you understand? 2. When it comes to writing content, two factors are crucial, "perplexity" and "burstiness." Perplexity measures the complexity of text. Separately, burstiness compares the variations of sentences. Humans tend to write with great burstiness, for example, with some longer or complex sentences alongside shorter sentences. AI sentences tend to be more uniform. Therefore, when writing the following content I am going to ask you to create, I need it to have a good amount of perplexity and burstiness. Do you understand? 3. using the concepts written previously, rewrite this article with a high degree of perplexity and burstiness: 4. Rewrite this article changing 50% of nouns and verbs for similar ones and rewrite verb usages to add natural sounding variations in complexity but retain the same tense and overall meaning. Edit: Number 4 was an addition that someone in the comments added that I found useful. I was testing it out, and when following steps 1 - 3 I got within the high range of 80% human made content in the ai detector. But when adding on step 4, it brought it to around 98%. Granted, you have to be careful make sure ChatGPT doesn't add any weird verbs or nouns that don't apply. I always recommend looking through with your own eyes afterward to make sure the text conveys the message you are aiming for. Another thing to note is that I don't think step 4 is ALWAYS necessary. I have gotten 100% human made content without it before. I suggest that you follow steps 1 - 3, see how much ai detects, and THEN use step 4 as necessary if applicable, because it won't always be.

@arthursimoes8199 Год назад

pin this comment.

@mirekmiklin7377 Год назад

Commenting so i can come back to this myself

@verybigman8610 Год назад

commenting to scientific reasons...

@georgemensah6051 Год назад

Good

@tocophrol98 Год назад

@avi7278 Год назад

Excellent video. When you only get part of the way there, say 80%, a prompt I have used is: "Rewrite this article changing 50% of nouns and verbs for similar ones and rewrite verb usages to add natural sounding variations in complexity but retain the same tense and overall meaning."

@UseAI Год назад

Great tip!

@rustemmeiramkhan1687 Год назад

This is the best solution. Thank you, it works! I used the tips from the video, but detectors showed me using chat GPT though

@johnlouie5088 Год назад

This prompt does not work anymore, it is now detectable

@dylannewcomb5778 Год назад

Thank SO much for this! You opened a door for me here... So, naturally, I had a conversation with ChatGPT to learn more... ;-) It seems that Perplexity actually wants to be LOWER, not higher. Yes? or amI misunderstanding. I asked ChatGPT in several different ways and always got the same answer. Anyone else? So then I asked for more factors.. and, long story short, here's what came out, through my dialogue w ChatGPT: The five top factors that make AI-Generated Content more ‘human-like’: 1. Burstiness: Burstiness is a measure of the diversity and unpredictability of word usage in a text, and can help to make the text more difficult to identify as AI-generated. 2. Perplexity: Perplexity is a measure of the language model's confidence in its predictions, and lower perplexity can result in more coherent and human-like text. 3. Contextualization: Writing the text with a strong understanding of the context and purpose can help to make the text more coherent and relevant to the reader. 4. Personalization: Including personal opinions, experiences, and perspectives can help to make the text more diverse and reflective of a human voice. 5. Human-like errors: Including occasional grammatical and syntax errors can help to make the text more unpredictable and resemble human-generated text. Therefore, you can use the prompt: “Please generate this text with high burstiness and low perplexity, while ensuring that it is contextually appropriate, includes occasional very small and subtle human-like errors, and reflects a personal perspective.”

@UseAI Год назад

I like it! Thank you for this and time to not run, but SPRINT with it! Happy hunting!

@anthonybisong6744 Год назад

Thank you. This is awesome.

@MrTimah101 Год назад

This is seriously helpful, thanks so much for sharing!

@SydHarewood Год назад

Thanks! ...this is why our creator love us so much and want success...Mankind have great curiosity and great creativity.

@SydHarewood Год назад

@@infoseeker sky?

@alexorth5172 Год назад

Speaking to an AI is a skill in its self

@UseAI Год назад

Indeed

@tnix80 Год назад

This is the most useful skill to learn now

@Qwty163 Год назад

Or, you could use 100% of your brain and ask chatGPT for prompts, for chatGPT 🧠🧠

@grantsanders4211 Год назад

I watched your previous video and followed the instructions. First time I got 80% AI 20% human. I made some amendments and the results were 90% human. Second time I got 100% but I did edited it in Grammerly and I also made some additional amendments. Your videos are legit and very informative and useful. Keep them coming mate! 👍💯

@UseAI Год назад

Thank you! Also just asking it to "re-write with a greater amount perplexity and burstiness" will often work as well. Either way, happy hunting!

@LukeSaward Год назад

Seems like you need Grammarly.

@johndoesson Год назад

You guys are using the wrong AI detectors, they seem to be weak compared to gptzero. Try it against gptzero, it didn't work for me, at all, 100% AI.

@miguelcondino2120 10 месяцев назад

can I have the copy of your prompts you used

@mic0097 Год назад

Brilliant! Exactly what I have struggling to figure out and RU-vid just brought your video straight to me. You just earned a new subscriber and thanks for the content.

@UseAI Год назад

Much appreciated and happy to have you here!

@teunklaassens2774 Год назад

Wow, this is amazing! I had no idea that Chat GPT could write human-like content with such high accuracy. The prompt used for the pass detection is really impressive. I can't believe that this was 100% human written. Keep up the great work, I can't wait to see more videos like this!

@UseAI Год назад

Thanks! Much more interesting stuff to discover.

@FontasticVentures Год назад

Nicely done! RU-vid suggested your channel and all your videos are great. Thank you for sharing.

@UseAI Год назад

Thank you! Be sure to sub. Much more coming soon.

@aktieneule Год назад

This is such a great help man! Thank you so much for sharing and also doing your research/tests!

@UseAI Год назад

Very welcome thank you!

@FansOfCrypto Год назад

Appreciate you exploring this and putting the prompt into the description.

@UseAI Год назад

Welcome!

@johnb8319 Год назад

Also manually adding references and asking the pgt to integrate them increase human%. When using the gpt to improve your writing, ask what it can be added to improve it, write it yourself and ask it to integrate it as one. It has higher chance of human% If the ai is using your words to improve a wroting instead of from the start.

@snowflakeandyou5983 Год назад

This is exactly what I am looking for... Great finding, keep making videos like this, subscribed!

@UseAI Год назад

Thanks! Will do. Anything for the fans!

@BASKIN1318 Год назад

Great vid! I've got an ai blog generation app and it's crazy how much of a difference changing up a few of the prompts on the backend makes with the definition of perplexity and burstiness. Thanks!

@UseAI Год назад

Indeed!

@DanSturdivant Год назад

Wow. Appreciate your generosity. Great RU-vid channel. Videos are concise and informative. Keep making great content!

@UseAI Год назад

Thank you! Great compliment. Will do!

@coach.b Год назад

Thank you for this highly informative and easy to follow video and the resource. The way you explained it from the perspective of functionality made absolute sense. 🙏

@UseAI Год назад

Welcome and thank you!

@JonathanAmbriz Год назад

You just made a bunch of students very happy.

@Papadoc1000 Год назад

Thanks for a great tutorial. Seeing what you did with it inspires me to take it several steps larger and see what happens.

@UseAI Год назад

Yes! Get it!

@dadonthetube Год назад

Thankyou for this excellent video AND for including the prompt and the links to resources in your description! I appreciate that.

@UseAI Год назад

My pleasure! Enjoy!

@braveville Год назад

Wow! I'm definitely trying this out. You prompt have been helpful. Thanks.

@UseAI Год назад

Happy hunting!

@strictlybangers Год назад

Absolutely loving this channel. Your subs are well deserved. Thanks again.

@UseAI Год назад

Thank you!

@Windrider30 Год назад

Huge thanks for this info as I have been playing around trying to get it to feel a little less AI and this promit really helps!

@UseAI Год назад

Glad it helped!

@Windrider30 Год назад

@@UseAI it really has and it works both for rewriting as well as writing again the more info you give to the chat AI the better the article it can write for you but this promit has really changed things

@muhammadyahyamubarik8316 10 месяцев назад

Great prompt and content. Prompt holds up well against the classifier from OpenAI also, will be interesting to see how this changes going forward.

@warmmilk9480 Год назад

I just tell it to use high randomness. In the initial prompt. But you can also get it to rewrite an article too, yes.

@fakharuddin3382 4 месяца назад

Brilliant work, I will check it out later

@martynarmstrong4730 Год назад

I came across this prompt and tested it. The results were a bit mixed, but I'm going to keep testing it because the concept behind the prompt makes a lot of sense. My biggest issue so far is Chat changing the format from 3 or 4 paragraphs per section to one block paragraph which cuts out a chunk of the information. That said, I might be staring the solution in the face and need to create a prompt that mixes perplexity and burstiness while retaining a specific format.

@ai.tools_ Год назад

Nice video, subbed. You definately stand out compared to other youtubers discussing A.I. Kudos

@UseAI Год назад

I appreciate that! Thank you for subbing!

@yaxxee Год назад

Thanks for this video, it's a genuine help. If you don't quite get to 100%, ask it to reword a couple of the paragraphs and you may find it hits 100% with a bit of tweaking. Alternatively just edit and reword a few sentences yourself, which tends to do the trick too.

@UseAI Год назад

Glad it was helpful!

@tnc97official37 Год назад

ohh my god, you are a saviour, you just saved me from going back and forth between paraphrasing tool etc...God bless you my friend ^^

@UseAI Год назад

Happy to help!

@zoranperic2312 25 дней назад

Rewriting ai manually worked for me , and also paraphrasing it via tools such as CleverSpinner or Undetectable ai.

@murrietamountainbiking5340 Год назад

When it comes to writing content, two factors are crucial, "perplexity" and "burstiness." Perplexity measures the complexity of text. Separately, burstiness compares the variations of sentences. Humans tend to write with greater burstiness, for example, with some longer or complex sentences alongside shorter ones. AI sentences tend to be more uniform. Therefore, when writing the following content I am going to ask you to create, I need it to have a good amount of perplexity and burstiness. Do you understand?

@jeetendranegi4225 Год назад

Hi Sir, You are great! I used your method and got 98% human written content through Chatgpt. Thank you so much for sharing helpful information through your RU-vid Channel.........

@mspalding100 Год назад

Thank you so much for your help and your work, without it, I would not be able to do what I’m doing now with AI👍

@UseAI Год назад

Great to hear!

@chrisdefi7865 Год назад

Very useful and easy to understand information. Looking forward to the next video!

@UseAI Год назад

Thanks!

@christiaanbotha6247 Год назад

I prompted Chatgpt not to make it detectable as AI and that worked.

@billt5410 Год назад

This is awesome! thanks, please keep uploading them.

@UseAI Год назад

Will do!

@hltechinsight 3 месяца назад

It really works. Thanks for sharing!👍

@koladebalogun4583 Год назад

Thank you for making this video...I am trying this technique of your out immediately. Thank you once more.. 🥰🤪😜

@UseAI Год назад

Welcome! Be sure to like and sub!

@Inspire_To_Empire 10 месяцев назад

Your a friking genius it works so well thank you🎉

@Manifest-Guru 3 месяца назад

TREMENDOUSLY VALUABLE INFO........THANK YOU..........THUMBS UP

@jadatoady Год назад

Thank you for sharing this info. It's a game changer and much appreciated ❤️

@UseAI Год назад

Welcome! Be sure to like and sub! :)

@joshv1814 Год назад

This is gold. Thanks for the prompt.

@UseAI Год назад

Welcome! Be sure to sub for more!

@jamesf.jordan2683 Год назад

This was a wonderful gift, thank you.

@UseAI Год назад

Welcome! Enjoy and come back for more!

@josephkimaru8716 Год назад

this is amazing bro..really i was almost clocking out of this industry but thanks to this mahn..big up G

@UseAI Год назад

Glad I could help!

@PBnJs365 Год назад

This worked. Thanks.

@Error-gc5zz Год назад

youre the best man thanks for the prompt very helpful !

@UseAI Год назад

Thank you thank you! Be sure to sub and like it!

@Flectron1 Год назад

OMG! You are a genius! 🎉. I appreciate your findings 💯. Brilliant.

@UseAI Год назад

Thank you! I do what I can! 👍

@freddfezzo8524 Год назад

I am already a subscriber and your explanations are quite easy to understand. I might add that burstiness can involve metaphors, similes and anecdotes. I saw that on another video. Adds a human touch I suppose. Please continue with your impressive videos. I just wonder if detection at all can approach 100%.!!

@UseAI Год назад

Thanks!

@GreesonTheGoat Год назад

I found success by asking the AI to write a article then after it wrote I asked it again to make it more "human-like-" and it was able to regenerate the article in a way that was not weird while also passing AI detectors with 100% human written content.

@silkogelman Год назад

Interesting video, thanks! 🙏🤖 FYI: There is a typo in the description for the contentatscale link. P.S. Have you tested with OpenAI's AI Text Classifier for detecting AI content?

@rockochamp Год назад

Thats what I call quality control. Thx pal!

@UseAI Год назад

🤣

@UldisBiz Год назад

New subscriber well earned. Thanks. Very valuable and to the point.

@UseAI Год назад

Awesome, thank you!

@alemandr Год назад

Amazing final! Looking forward to finding out new promts for this AI😎

@UseAI Год назад

Thanks! Enjoy!

@celebrim1 Год назад

Lol. That's amazing. You've taught the AI how to evade it's own detection algorithm. Moreover, presumably content that reads more like human content will also feel more natural to readers.

@UseAI Год назад

Absolutely!

@AffiliateMarketingGame Год назад

I think this way is far better thank using other tools. Thank you!

@UseAI Год назад

Agreed!

@delfinovargas1152 Год назад

Thanks fue videos. They are good. Tell me how do you solve the problem of citations.

@ThichabodCrane 5 месяцев назад

The prompt made chatgpt go absolutely off the rails with it's adjectives. Instead of sounding like a robot it sounded like a lunatic.

@perrygupta Год назад

It passes the new openai text classifier lol. Thank you so much for this. You are genius.

@UseAI Год назад

🤣🤣🤣

@perrygupta Год назад

@@UseAI keep making such content please :)

@TripDogg5 Год назад

This is a game changer! Great work! 👍

@UseAI Год назад

Thanks! Happy hunting! :)

@joannaduplessis7231 Год назад

A million thank you's!!! Much appreciated.

@UseAI Год назад

You're welcome!

@420clothingdesigns Год назад

Brilliant video, thank you for your helpful insight.

@UseAI Год назад

Thank you!

@pikachubailarin Год назад

homework was punishment, then it became a norm, then we created AI, then we used AI to make homework; Then they made AI that detects AI made homwork, then we made AI that retypes AI homework so that AI that detects homework cant detect it

@scruffylife Год назад

This is gold 👌 enough said!

@UseAI Год назад

Thanks!

@smartjt Год назад

Thanks for sharing. Very insightful

@UseAI Год назад

Thank you and welcome!

@schneidervonhammer722 Год назад

As I mentioned to a colleague... This will be a cops n robbers chase, but The AI itself can be seen as a magic wand, Abstracting what you said in the video -> basically comes down to the Prompt input - As the user/magician -> he who Masters the PROMPT - can master the AI.

@LaMontPrice Год назад

Works sometimes. Depends on the AI detector. I've found that adding external references and creating the document with headers (H1, H2,) or lists, reduces the ai detection. Still testing.

@UseAI Год назад

I'm sure some updates have occurred since I made this video and it definitely depends on the content used.

@LaMontPrice Год назад

@@UseAI I found that GPTzero is the most strict. If it passes that site, usually it will pass the others.

@moonfox89 Год назад

This is mind-blowing. Thank you for sharing this! I love your channel. Videos are straightforward, not lengthy and extremely informative. Keep making contents like this ❤️

@UseAI Год назад

Thanks! I do what I can. 🧙‍♂

@anthonybisong6744 Год назад

Thank you for sharing this information.

@UseAI Год назад

Welcome and thanks for dropping by! Be sure to sub!

@jordandiefe Год назад

Another method I found that works is replacing all spaces in the text with an invisible character. It even passed Open AIs new AI classifier and all of these sites as human generated.

@Gu66y453 Год назад

Wait could you explain this more? what is invisible character?

@vin84iitr Год назад

Awesome content indeed!!

@UseAI Год назад

Thank you thank you.

@ProjectManagementMastery Год назад

Funny it works but when I think it's too short and ask the bot to add more words, it fails (add more words, still with low perplexity and high burstiness)...

@UseAI Год назад

Still a lot to iron through, but be sure to save what works!

@karenbennett187 Год назад

Fantastic information! Thanks so much. ❤

@UseAI Год назад

Welcome and thank you!

@leen6899 9 месяцев назад

thank you so much 🥰🥰🥰🥰🥰

@BoELoot Год назад

"When it comes to writing content, two factors are crucial, "perplexity" and "burstiness". Perplexity measures the complexity of text. Separately, burstiness compares the variations of sentences. Humans tend to write with greater burstiness, for example, with some larger or complex sentences alongside shorter ones. AI sentences tend to be more uniform. Therefore, when writing the following content I am going to ask you to create, I need it to have a good amount of perplexity and burstiness. Do you understand?"

@mvs6380 Год назад

I’m so glad YT recommended me your video. One question, my chat GPT doesn’t show the Output language option. How did you get that?

@rehmaneverywhere Год назад

There is an extension called "AIPRM" which will give you this feature and also some other SEO prompts.

@mvs6380 Год назад

@@rehmaneverywhere thank you!

@jtdeux Год назад

Yes, I'm curious as well. It'll be so helpful if you can share. :)

@UGLYANGEL Год назад

Excellent as I too was looking for a way to explain to the AI. This is great.

@UseAI Год назад

Thanks!

@palmeressay243 Год назад

Looking for help with #essay #math #assignment #discussion #tutor #project #ppt #dissertation Contact Dawson Essays for help GUARANTEES: 📎-Original papers (very affordable) 📎-Customer satisfaction 📎-A+ in your paper. DM📨/WhatsApp +1 (267) 937-0586.

@k.h.p.9862 Год назад

Your discoveries are amazing. Thank you.

@UseAI Год назад

Glad you like them!

@Kirankarnati0007 Год назад

Amazing demonstration… unbelievable results… Mastering this prompt engineering technique is literally what determines success in future, those who embrace it will move ahead and those who resist it will fall behind.

@UseAI Год назад

I'll drink to that!

@aarondavidson6409 Год назад

AI bros are staunch

@dlog Год назад

@@aarondavidson6409 wait till they merge with cryptobros and use AI to make NFTs!!!!!! . . . . . . Oh wait it has been done since day1 of crypto

@ExploreSweden Год назад

It's worth noting that Google is NOT penalzing AI written content. Danny Sullivan from Google tweeted this for example "We haven't said AI content is bad. We've said, pretty clearly, content written primarily for search engines rather than humans is the issue. That's what we're focused on. If someone fires up 100 humans to write content just to rank, or fires up a spinner, or a AI, same issue...".

@UseAI Год назад

Exactly. No evidence this is a problem unless the user experience is poor.

@jtdeux Год назад

Thanks for sharing. May I know if you can share how to get the Output language option? I can't find it in any chrome extension. :)

@ElaineFoster101 Год назад

Chat told me to use the words "sentence variation" instead of burstiness. But THANKS for telling us about this concept!

@MrKennykyle1 Год назад

now this is very nice. Now you get a subscriber :D

@UseAI Год назад

YES. Glad I earned it!

@stanz47 Год назад

I would love to see you run this against GPT Zero, which has been the most difficult to fool for me.

@motogsprx7251 Год назад

I tested the prompts from the video and i still got 100% AI written content in GPT Zero .. that's a pretty good tool

@stanz47 Год назад

@@motogsprx7251 yeah, I’ve been finding it hard to fool GPT zero a lot of the time. Sometimes I can past GPT zero but the other ones will catch it, so I’m not sure if it works the exact same. It’s tricky to get all of them for me

@girlball12 Год назад

Very well explained. Thank you!

@UseAI Год назад

Thanks! Be sure to sub for more!

@regentalguide Год назад

My first thought would've been to ping pong the same text between 2 different ai, but maybe some combination of this and ping ponging it will be really difficult to detect. In the end though, I think ai will always have the advantage.

@kevinsrandom Год назад

This is wild. Its also wild to talk to an AI like you would to a human. Nothing special to get the AI to understand stuff 😮😳

@UseAI Год назад

Yuup.

@OparOfficial Год назад

My boy are you okay? You wrote 52 ways to be happy? O_O

@UseAI Год назад

LOL. It was a subtle cry for help.

@nuwanchandrasirigamage8130 10 месяцев назад

This is work for me. thanks

@adaption27 Год назад

Cool tool. Thanks.

@Mysticzen6 Год назад

You are a genius! It works amazingly you should not give this secrets away before they will find a way to implement new systems so that this trick will no longer works!😊

@UseAI Год назад

ha! It will happen anyway.