[021] ChatGPT Spam Filter [Deleted]

[021]

I think your maths are a bit off as $0.002 is per 1000 tokens not per token

• Total tokens per day = 390 tokens per post * 200 posts per day = 78,000 tokens per day
• The total number of tokens used over a 30-day period would be 78,000 tokens per day * 30 days = 2,340,000 tokens
• Based on the current pricing for the ChatGPT GPT-3.5 Turbo model of $0.002 per 1K tokens, the estimated cost for 2,340,000 tokens would be = (2,340,000 / 1000) * $0.002 = $4.68 for 30 days

What a "genius" I am... I also thought when I wrote the message "Something is somehow expensive, it can't be like that", but I still don't understand where the mistake is...
Thanks a lot. I'm willing to give you a discount for this if you'd like to purchase the add-on

 

[eva2000]

Actually based on https://openai.com/pricing

Multiple models, each with different capabilities and price points. Prices are per 1,000 tokens. You can think of tokens as pieces of words, where 1,000 tokens is about 750 words. This paragraph is 35 tokens.

That comes to average number of tokens per character to be around 0.11 to 0.14 tokens per character

390 tokens per post / 0.11 to 0.14 tokens per character = 2,786 to 3,545 characters per post

Assuming 200 posts per day for 30 days, the total number of characters for the entire period would be:

Total characters for 30 days = 200 posts per day * 30 days * 2,786 to 3,545 characters per post = 16,716,000 to 21,270,000 characters

Total tokens for 30 days = 16,716,000 to 21,270,000 characters * 0.11 to 0.14 tokens per character = 1,838,760 to 2,956,980 tokens

Based on the pricing of ChatGPT GPT-3.5 Turbo model of $0.002 per 1K tokens, the estimated cost for 1,838,760 to 2,956,980 tokens would be:

Total cost = (1,838,760 to 2,956,980 / 1000) * $0.002 = $3.68 to $5.91

and yes I actually asked ChatGPT for these calculations

 

[021]

Actually based on https://openai.com/pricing


That comes to average number of tokens per character to be around 0.11 to 0.14 tokens per character

and yes I actually asked ChatGPT for these calculations

Hahah, nice

 

[eva2000]

Hahah, nice

And remember ChatGPT API token calculations are for both input + outputs. So imagine if spammer posts a 5000 word essay LOL

 

[Spartan]

It'd be great to get this to find duplicate users based on common word use by multiple posters. After enough posts most users create a pattern on using certain words more often, misspelling the same word multiple times, repeating common grammar mistakes, etc.


Example: Enter a single username, then search for words used most often (limit for x lettered words or higher), and words repeatedly misspelled. Then search data base for matches. May be limited to common ip addresses or isp.

 

[021]

021 updated [021] ChatGPT Spam Filter with a new update entry:

1.1.0

Added an option that allows you to choose what to do when spam is detected in a message - send for moderation or reject.
Addon renamed to [021] ChatGPT Spam Filter.
The minimum version [021] ChatGPT Framework for this version to work has been increased to 1.2.0+.

Read the rest of this update entry...

 

[Mr Lucky]

Will this detect if a (human) spammer has posted content that was generated by ChatbotGPT?

 

[021]

Will this detect if a (human) spammer has posted content that was generated by ChatbotGPT?

Give me an example of such a text and we will check it. It is important to note that ChatGPT is configured to not generate spam content.

 

[Mr Lucky]

Give me an example of such a text and we will check it. It is important to note that ChatGPT is configured to not generate spam content.

I will look out for it (I think there was one on a forum a I moderate but got spam cleaned)

I was thinking about this thread:

Starting to see spam registrations and spam postings, where the content is very much like ChatGPT phrasing and language.
I feel this is going to bring a new wave of spam methodology.

Have you tried fighting ChatGPT with ChatGPT?

 

[Mendalla]

Give me an example of such a text and we will check it. It is important to note that ChatGPT is configured to not generate spam content.

Right, but someone could get it to generate seemingly innocuous content, copy and paste to a post, and then embed spam links in strategic locations. Much like a human spammer does now with edited boilerplate text.

 

[Kirby]

the add-on sends the text and other information about the post

What exactly is this other information, does that include data like Username, E-Mail, Useragent or IP-Address?

 

[021]

What exactly is this other information, does that include data like Username, E-Mail, Useragent or IP-Address?

Username and number of registration days

 

[DiscussionBucks]

This seems to be a good addon idea. I do have a question though. Does this addon help detect AI-generated content? Unfortunately, Google has started marking AI-generated content as spam which could cause some long-term SEO trouble. If this addon helps detect AI-generated content I am definitely going to purchase.

 

[021]

This seems to be a good addon idea. I do have a question though. Does this addon help detect AI-generated content? Unfortunately, Google has started marking AI-generated content as spam which could cause some long-term SEO trouble. If this addon helps detect AI-generated content I am definitely going to purchase.

No, as long as the content generated by the AI is not spam, the add-on will not block it. This doesn't sounds truly and I have not been able to find information that Google does such things. Where did you find this information? According to my information, Google has taken an adequate position on this question: if the content is of high quality, it will not receive a penalty, if not, then, in fact, like any other low-quality content, it will not play in favor of SEO

 

[JordanH]

This seems to be a good addon idea. I do have a question though. Does this addon help detect AI-generated content? Unfortunately, Google has started marking AI-generated content as spam which could cause some long-term SEO trouble. If this addon helps detect AI-generated content I am definitely going to purchase.

I second this.
Would it be possible to modify the addon or create a new one that does detect content being generated by AI?

 

[Alpha1]

Its undesirable if random members would be able to post AI generated content because so far only spammers would do that to look like real users. In this respect AI content is a relatively new attack vector to defeat content moderation. Having AI detection would be useful to have to counter this.

 

[JordanH]

No, as long as the content generated by the AI is not spam, the add-on will not block it. This doesn't sounds truly and I have not been able to find information that Google does such things. Where did you find this information? According to my information, Google has taken an adequate position on this question: if the content is of high quality, it will not receive a penalty, if not, then, in fact, like any other low-quality content, it will not play in favor of SEO

Google Says AI Generated Content Is Against Guidelines

Content generated automatically using AI writing tools is against Google’s Webmaster Guidelines, says Search Advocate John Mueller.

www.searchenginejournal.com

For us these would, essentially, still fall into the category of automatically generated content which is something we’ve had in the Webmaster Guidelines since almost the beginning.

And people have been automatically generating content in lots of different ways. And for us, if you’re using machine learning tools to generate your content, it’s essentially the same as if you’re just shuffling words around, or looking up synonyms, or doing the translation tricks that people used to do. Those kind of things.

My suspicion is maybe the quality of content is a little bit better than the really old school tools, but for us it’s still automatically generated content, and that means for us it’s still against the Webmaster Guidelines. So we would consider that to be spam.

 

[Forsaken]

Google Says AI Generated Content Is Against Guidelines

Content generated automatically using AI writing tools is against Google’s Webmaster Guidelines, says Search Advocate John Mueller.

www.searchenginejournal.com

Watch that change when they make Bard public .

 

[ProCom]

Google Says AI Generated Content Is Against Guidelines

Content generated automatically using AI writing tools is against Google’s Webmaster Guidelines, says Search Advocate John Mueller.

www.searchenginejournal.com

That's an older article. I believe there is newer / more-recent updates to this, including Official comments from Google

Google Search’s guidance about AI-generated content | Google Search Central Blog | Google Developers

In this post, we'll share more about how AI-generated content fits into our long-standing approach to show helpful content to people on Search.

developers.google.com

Specifically:

"Is AI content against Google Search's guidelines?​

Appropriate use of AI or automation is not against our guidelines. This means that it is not used to generate content primarily to manipulate search rankings, which is against our spam policies.

Why doesn't Google Search ban AI content?​

Automation has long been used in publishing to create useful content. AI can assist with and generate useful content in exciting new ways."

 

[DiscussionBucks]

It appears as if AI-generated content is quite controversial in terms of SEO. With that in mind, whether it is against search engine guidelines or not, I still feel like determining whether something is AI-generated or not would be a great addon to have for forum owners who would prefer such content not be posted on their platform.