You are viewing a single comment's thread from:

RE: Grapes ๐Ÿ‡ : A Taste of History in Every Bite ! Health Benefits ๐Ÿ‡

in Hindwhale Community โ€ข 2 years ago

Testing this article on the GPT-2 Output Detector gives a 99% chance that it has been generated by ChatGPT.

This would suggest therefore that this is not your own original work.

Is this your own original work @lavanyalakshman ?

If so you might want to cite your sources.

Passing off AI generated text as your own is as bad as plagiarism.


cc @shravana @damithudaya @jyoti-thelight @deepak94 @aaliarubab @pathanapsana @pea07 @abuse-watcher


Community Curator Team 3 - remember you should all approve the selected posts of the week. It is surprising that none of you spotted that this article has been generated by ChatGPT.

@goodybest @graceleon @damithudaya @afrizalbinalka
@wilmer1988 @sualeha @f2i5

Sort: ย 
Loading...

This is a great link! Thank you! I hope other curators and moderators see that, too. Before now, all I had was this, which is somewhat hard to interpret...

I am aware of some accounts that I have long suspected of AI content generation, but I could not be confident whether they were AI-generated or English as a 2nd language until now.

I just ran a handful of their posts through the GPT2 detector link from your comment, and it agrees that they're most likely inauthentic (i.e. 98-99+% labeled "Fake"). If I'm correct, the amount that they've collected over the course of a couple years is not trivial.

Are you interested in the list? If so, how can I send them to you?

ย 2 years agoย 

It might be best to pass it on to the AbuseWatcher team in the first instance.

Or you can use their new !report feature in a comment.

Thank you.

Hello @remlaps
Don't hesitate to give us the information. Please report to our Discord server or use the !report command.

Loading...

Greetings steemit team, in also name of the whole team we apologize and we will take as usual the necessary actions to improve and not make this mistake again.

ย 2 years agoย 

Greetings Steemit Team! We will be more careful in future about the detection of ChatGPT Content. We should team apologize to you.
Ok from now on we add one thing more in our checking criteria , ChatGPT content detector. Today, will publish an announcement and guide my team to be more careful in future.

https://www.zerogpt.com/

We will detect in future.

Thank you very much for letting us know.

ย 2 years agoย (edited)

Thank you very much for bringing this matter to our attention. Lavanya, who has a reputation and great responsibility and a lot of power, is unlikely to do like this. Maybe she uses a translator or a spell checker tools like "Grammarly". Because now Grammarly software uses AI method to convert the words itself. Also, the tool you mentioned is not accurate, shows the sentences normally typed by humans as fake.

For example, I Checked above phrase that I written , it shows as 64% fake, what to say..

image.png

ย 2 years agoย 

Well, it was my mistake to select this for curation and as the best pick. I apologize for it on behalf of the team.

Normally I check posts using an AI detector, anyhow this has been missed. It will not happen again and we will be more careful and thorough during the curation.

Thank you very much for enlighting us on this matter, and I was wondering about a recommendation for a good AI detector.

So, this, https://openai-openai-detector.hf.space/, will be taken for future works.

Best Regards!

ย 2 years agoย 

Hello dear @steemcurator01 thank you so much for spotting this abuse, but I want to apologize in behalf of my team, it's our bad ๐Ÿ˜” and we are sorry, we hope to do more better. It just sad that someone with high reputation like @lavanyalakshman could do this, we will never trust anyone next time, I just wonder as an admin what message he's passing out to others. Moreover, I just learnt about chatGpt not too long ago from @jesusjacr post, yeah just a day I even copy out the detector link and kept it safe, but we promise to do better next time. Please We are sorry ๐Ÿ™ and thank you for understanding.

Thank you for informing us Mr @steemcurator01 about the implementation of ChatGPT or commonly called chatbot.

From several sources of articles that I have read that this application cannot be used all of its contents because there are several sentences taken from the article that are converted into other sentences. And this I once tried to learn more about AI applications.

Even some specialised plagiarism applications that I used were not detected accurately. But I have my own way to detect sentences.

Screenshot_2023-03-03-04-20-09-47_40deb401b9ffe8e1df2f1cc5ba480b12.jpg

this article was detected using chatbot/chatGPT

In this case, we will be more vigilant about articles published by authors on the Steemit platform to avoid using AI applications. Thank you for letting me/Team 3 know about this.

Cc @graceleon @irawandedy @damithudaya @wilmer1988 @sualeha @goodybest

๐Ÿ˜„๐Ÿ˜„๐Ÿ˜๐Ÿ˜๐Ÿ˜

Screenshot_20230303_115818_Chrome.jpg

ย 2 years agoย (edited)

Means most of them it shows as fake ๐Ÿคฃ
@steemcurator01 , Are you using ChatGPT for generating this comment? I don't believe It's a tool; it doesn't give 100% original results.

If you believe, Tools on judging Me, then your wish

I'll let ChatGPT write my thoughts for me...

The quality of English in the comment may not be consistent with the quality of English written in the article because the comment could have been written by someone who does not have a strong command of the English language.

Coin Marketplace

STEEM 0.21
TRX 0.20
JST 0.033
BTC 94309.20
ETH 3080.15
USDT 1.00
SBD 2.99