Shall I be downvoting posts made by AI tools?

onealfa in #ai • last year

I asked the above question (in the title of this post) on ChatGPT.
Here is the answer I've got:

It is important to consider the potential impact of downvoting posts made by AI tools on the future of social networks:

1-Encouraging authenticity: Downvoting posts made by AI tools can be a way of encouraging authenticity and human interactions on social networks. This can promote a culture of genuine engagement and build trust between users.

2-Discouraging unethical practices: Downvoting AI-generated posts can be a way of discouraging unethical practices such as using automated tools to create content. This can help create a level playing field for all users and promote fairness and equity.

3-Upholding terms of service: Downvoting AI-generated posts can be a way of upholding the terms of service of social networks, which typically prohibit the use of automated tools to create content. This can help ensure that users abide by the rules and guidelines of the platform.

4-Promoting innovation: Downvoting AI-generated posts can be a way of promoting innovation and creativity on social networks. This can encourage users to come up with new and unique content ideas that are not reliant on automated tools.

In conclusion, whether or not to downvote posts made by AI tools is a personal decision. However, downvoting such posts can potentially encourage authenticity, discourage unethical practices, uphold terms of service, and promote innovation on social networks, which can all contribute to a better future for these platforms.

As you can see, 99% of this post is created by AI, is clearly marked as so, and is quoted to stand out.
It is now your task to decide - to upvote it, to downvote it, to ignore, to support, or to argue.
Or, perhaps, join me in my efforts. I intend always downvote posts, when I have a clear evidence of hidden AI use.
Because, IMHO, posts made by AI tools, without a clear indication of AI use - is nothing more but an easy way of STEALING rewards from the honest creators. Period.

@onealfa
a.k.a. Toruk

#leofinance #chatgpt #abuse #hive #openai #vyb #proofofbrain #curation #rewards

last year in #ai by onealfa

48.88 VYB

Sort:

Trending

[-]

ifarmgirl-leo last year

AI even knows about authenticity and ethical practices. DV it is!

Posted Using LeoFinance ^Beta

0.00 VYB

[-]

cryptoandcoffee last year

Upvoted as you are raising serious concerns we all have and I will be downvoting as I clearly see it as a form of abusing what we have here. People talk about enhancing their posts blah blah but why would you unless you are trying to game the system. research is one thing but writing the post is a clear no in my book. We know many try and cheat the system and we cannot let them loose with a loophole like this as Hive will be a mess in no time.

0.00 VYB

[-]

leonmaster last year

I am personally convinced that we must advance at the same time as new technologies are developed, of course, the challenge is that we can find all the technologies that in some way "derive from the internet" (cryptography, IA, blockchain, metaverse...) correct, responsible and reasonable use.

I invite you to see my last publication in which I address this topic. (comments are welcome).

0.00 VYB

[-]

virtualgrowth last year

Maybe in time AI will have better (placed) tools that will help content creators. This could be with art, pictures, research. Perhaps it is a collaborative work with the AI and the human along with the machine(s). Important to state that AI is being used.

Posted Using LeoFinance ^Beta

0.11 VYB

4 votes

[-]

hive.defender last year

Hi there, as our parting gift I will share with you three users you should check for AI, we will still report them to HiveWatchers, good luck downvoting AI post.

theelectlady
imam-samudra
putroe-hive

0.00 VYB

[-]

poshthreads last year

https://leofinance.io/threads/@magnacarta/re-leothreads-tmupj
_{The rewards earned on this comment will go directly to the people ( magnacarta ) sharing the post on LeoThreads,LikeTu,dBuzz.}

0.00 VYB

[-]

dbooster last year

Downvote obvious and hidden use, obviously. But I think we do have to be careful about false positives. As AI gets better, as we gain the ability to train AI on our own writing so that it adopts our writing voice, it will become more difficult to detect and false positives will increase. I mean just as an example, I've been writing online for over 20 years. Feed all that writing into an AI and I imagine it would be able to do a pretty convincing job of sounding like me. If and when that is possible, it will become harder to detect, and i I imagine a lot of people who are actually writing their own will start being flagged.

So yes, I completely agree with you in theory. But I do think we need to be careful. I want to DV people who are trying to cheat the system, but I don't want to DV people who were caught as a false positive and don't deserve it.

0.12 VYB

3 votes

[-]

onealfa last year

...we do have to be careful about false positives...

I much agree with you - not a very easy, nor a simple task is raising in front of us

Posted Using LeoFinance ^Beta

0.00 VYB

1 vote

[-]

anna-newkey last year

Would we always know it's been created by AI?
Will it be highlighted as in your excellent post?
I did read this as your thoughts, until I came to the last paragraph...
Most people wouldn't have a clue, how could they?
Thanks for the question! So far I've not downvoted anything, and I had no intention to, but this might be a valid reason to use it... if I know it really is AI and used abusively.

0.00 VYB

[-]

anna-newkey last year

On second thought, should it be used at all?
It makes it much harder for those of use who still type up posts and that might make Hive unattractive, something that nobody wants.

0.00 VYB

[-]

hivebuzz last year

Congratulations @onealfa! You have completed the following achievement on the Hive blockchain And have been rewarded with New badge(s)

You distributed more than 210000 upvotes.
Your next target is to reach 220000 upvotes.

_{You can view your badges on your board and compare yourself to others in the Ranking}
_{If you no longer want to receive notifications, reply to this comment with the word STOP}

Check out our last posts:

The Hive Gamification Proposal

0.00 VYB

[-]

acesontop last year

I upvoted before actually reading it but I stand by such attitude. Using AI one could create a couple of accounts and start flooding Hive with AI created content which can, as mentioned by you, take rewards from the same pool as honest content creators. From what I know there will be a bot detecting such content.

0.00 VYB

1 vote

[-]

jfang003 last year

I wouldn't downvote this post because you mentioned directly that it was made by AI. I think it's about whether or not people know it is AI created or not. If someone conceals this fact, then I think they might be feeling guilty about their actions.

Posted Using LeoFinance ^Beta

0.00 VYB

2 votes

[-]

stayoutoftherz last year

I agree, the crucial thing indeed is without a clear indication of AI use.
If it is disclosed, the reader can still see a value or not in the post, like in any other post, as if other tools were used. But that would not be a reason for downvoting (regardless if 49% or 51% AI usage, like that silly treshold from hivewatchers).
But even if it is disclosed, if a user would churn out 5 posts a day with predominant AI usage and no other intrinsic value, e.g. just a pic and nothing else, that would be a different story, wouldn´t it?

0.00 VYB

[-]

onealfa last year

...if a user would churn out 5 posts a day with predominant AI usage

Abuse has many different forms. One of them is SPAM

Posted Using LeoFinance ^Beta

0.00 VYB

1 vote

[-]

libertycrypto27 last year

If AI itself recommends downvoting posts written by AI the choice to downvote them is absolutely right.
You thought of a really brilliant post asking that question to the AI! I will forward this post of yours to anyone who tells me that AI on Hive is good! !LOL
I will never vote for a post written by an AI and I will never use AI to write my posts because I enjoy writing and I would be a fool to take away a fun time. I prefer human emotions and even human mistakes because beauty is often in imperfections and not in robotic, cold perfections

Of course I voted for this post but because your idea and conclusions even if they are 1% of the post are worth much more than the 99% written by AI!

Posted Using LeoFinance ^Beta

0.05 VYB

3 votes

[-]

lolzbot last year

lolztoken.com

I'm on a seafood diet.
When I see food I eat it.
_{Credit: happyme}
@onealfa, I sent you an $LOLZ on behalf of @libertycrypto27

(4/10)

PLAY & EARN $DOOM

0.00 VYB

[-]

magnacarta last year

AI itself is just a tool, like a hammer or a gun or a car. It is neither good nor bad; it simply exists. What matters is the intent of the person using the tool (in this case, AI).

If AI is used as a supplement to whatever we produce, that is OK. If AI is used as a substitute for our work and effort, that is not OK.

AI which deals with a specific task-- grammar checking, language translation, medical diagnosis, ...-- is OK since it isn't used to replace our own creations.

The problem comes when AI is used to replace our creation ("ChatGPT, tell me why HIVE is superior to USD").

Many people treat AI as the final authority on whatever topic is presented to it, and that just isn't true. Even if we choose to use ChatGPT-style AI-- since that is where most of the problems originate-- we can still override its output when we create our content.

I think the worst aspect of using a ChatGPT-style AI is how easy it is for people to deceive readers into thinking the content was created by a human being when in fact it was generated by the AI based on a prompt. It's the deception which is at the heart of whatever anguish and anger and uneasiness we have when we discover that ChatGPT-style AI was used to produce the content.

Perhaps this comment should have been made as a separate post; it's not the first time that's happened, and it won't be the last. I'm still trying to sort out some things before I can publish a long form post on this topic. I was just trying to contribute something new to the discussion.

At least for now, I will take it on a case-by-case basis. When use of ChatGPT-style AI is declared and noted, I reserve the option to upvote. When it's done in a stealth manner and I can detect that, I will downvote (such as it is for me).

Posted Using LeoFinance ^Beta

0.00 VYB

2 votes

[-]

dagger212 last year

Yes, it seems to me intent should really be the deciding factor. If the "author" sources the AI content (as you did here), then writes their own conclusions based on that, then I have no problem upvoting the article. If they are using AI to simply post an article to get upvotes without acknowledging it's use, that is just trying to game the system and probably deserves a downvote. If they are somewhere in the middle, just ignoring it is probably the way to go. AI is here, that fact is irrefutable. It will be up to each of us individually to decide how we want to use it, how to reward it, or whether it should be flagged. There is no right answer.

I don't think the biggest problem is going to be these articles being generated. In my opinion, the biggest problem will be when they start feeding each other with circle jerks of AI voting. That is what needs to be uncovered and stopped as soon as it appears.

Posted Using LeoFinance ^Beta

0.12 VYB

2 votes

[-]

muratkbesiroglu last year

I worked in machine learning for over ten years during my banking career. Starting in 2016, I wrote novels and stories about artificial intelligence. These books were published in Turkey and reached science fiction enthusiasts. Therefore, I had the opportunity to think and work on the subject's theoretical and practical aspects.

In my opinion, we should not consider whether the content is produced by artificial intelligence or human beings as a criterion. We should evaluate the content produced according to its quality. Because after a while, we will not be able to understand whether a post was produced by artificial intelligence or by a human. The use of artificial intelligence devalues the efforts of content creators, but this is an inevitable process. Bans may delay the issue for a while, but if AI does the job better, we should let it do it.

On the other hand, the content produced by artificial intelligence is not usable in its current form. Those who have in-depth knowledge of the content of the content detect any inaccuracies in the content produced by artificial intelligence. Also, how it works makes it difficult for AI to list the resources it benefits from.

The content produced by artificial intelligence is not much different from automatically generated statistical reports. The curation process of these reports was also controversial. Considering that creating and maintaining these reports takes resources, I think they deserve an upvote.

How can we describe the content produced by artificial intelligence? Various stages of the process may be automated. Will we accept ninety percent automated content but reject completely auto-generated content?

I think we should give upvotes to the content produced by artificial intelligence in proportion to the value it creates. However, we need to consider some ethical codes here. If the content is produced by artificial intelligence, we should know it. Of course, we can downvote if the person using artificial intelligence tries to convince us that he/she produced the content manually. Just like in the images, showing the artificial intelligence algorithm as a source would be appropriate.

On the other hand, as Hive curators, we may encourage content creators to use artificial intelligence tools because some of the best content is produced by human-machine collaboration. Newly developed artificial intelligence tools allow us to access information more efficiently than search engines. We can use artificial intelligence for general information and enrich the content by adding personal comments.

Note: I wrote all the sentences in this comment manually :)

0.79 VYB

5 votes

[-]

vimukthi last year

On top of this, there is the probability of false positives and false negatives. It is best to focus on the value of the article itself. The AI content is full of flaws (especially when it comes to factual accuracy). If one type of content becomes easily made by AI, the overall value of that content itself will go down.

This could lead to make personal content and personal style more valuable and rare. The best response to technology is adopting to it instead of becoming mindless drones or overly conservative. Even if HIVE keeps AI, away, rest of the internet will be flooded by AI content. The best thing to do is focus on the type of content AI cannot produce right now effectively.

It will be an arms race and I am ready for it.

Posted Using LeoFinance ^Beta

0.00 VYB

[-]

mypathtofire last year

I understood that Hive was about Proof of Brain and not Proof of AI. The hidden AI use is the worst part for me.

0.00 VYB

2 votes