Press the Any Key

avatar

One of my colleagues and I were playing with ChatGPT yesterday, to see what are the capabilities for it from a work perspective. For example, we asked it to tell us about the company we work for. We asked it to give us reasons why what we do for customers is important. We asked it to write a sales pitch. Overall, it did a great job and would make creating a base easier for basics, but...

It is definitely plagiarism.

While it was able to write in natural language well, it was also taking content almost directly from available marketing material and content that both of us know quite well. In fact, I have worked on some of the content and there were slices of my sentences in there also. So, what it is doing is spinning the content already available, and it isn't asking if that is okay to do - especially for commercial usage.

P2120439 (1).jpg

Spinning and it isn't just frowned upon on Hive, it is discouraged heavily through downvotes, as is of course plagiarism, so why would AI created content that automatically plagiarizes and spins content be treated differently from if a person does the same?

Some people seem to think that AI content isn't an issue and should be rewarded, but how would you feel if for example, my content was generated through AI and I was getting rewarded for it as I do? Or @taskmaster4450's, or any other well-rewarded content creator? What if you knew that I was using AI as a base for my work and instead of taking hours to create a single piece, it was taking me ten minutes?

Happy with that efficiency?

Of course, while I don't know about others, I am definitely not doing this because not only is AI generated content plagiarism, it is also not mine. It would be interesting to see what it would create if it only had access to my work, as I have almost 6000 posts written hudreds of topics on HIVE, that equate to something like 30x 300-page novels, plus comments. What would an AI create?

Would it still be mine?

I don't think so.

And, the reason is that even if it is writing a new article using my voice, it isn't using my experience garnered between then and now. Since we are always changing, it won't actually represent my opinions in the moment, even if it is close.

People don't have an intuition for this, especially those who are unable to produce as good quality content as the AIs. No one is going to steal the work or spin the content of low-quality, low-experience, content creators, are they? Yet, most people don't want their own work stolen, especially for gains that they get no part of. We all have experience with those who take our ideas and present them as their own, and I am yet to meet someone who likes it happening. Just because the content comes from sources off Hive, it doesn't make it okay.

And with an AI, how do you know?

The AI isn't citing sources, it is just trawling, collecting and combining. In the case of testing with my colleague, the only way we knew where it came from, was because of our very direct relationship with its creation, and the people who originally wrote it.

It raises moral questions about those who use it for reward, doesn't it? Even using it as a base, there is so much room for content theft and of course, the easier it is, the more it happens. The first article used it as a base, the second spun it, the third just cuts and pastes in entirety.

Are you still a content creator?

Your content has value.
Your voise has value.

Where is the line start of what is yours, and what belongs to the unseen and unreferenced voices of those who originally created the output?

And if it isn't your voice, people shouldn't be providing value to you. It isn't just citing the usage of AI either, because that doesn't reflect the actual sources of the information provided.

And for a lot if the people using AI to create their content, it would appear that they themselves have little to no skill or experience in these areas, so how clever can the usage of the tool be and, what can they add in value through comments?

Google answers, no doubt.

What I like about the usage of AI, is it is raising questions of what is valuable, something that has to happen going forward. Ont top of this, it is also raising the ownership equation and in an age where content source is rewarded, this is going to provide much needed usecases for blockchain, through content verification and proof of human.

These are key to blockchain adoption and require careful consideration, but there is already a massive need for being able to trust content and verify source, as well as legitimacy. So far, the algorithms are providing response based on relevancy, but that doesn't mean what is provided is trustworthy and instead, tends to feed a process of confirmation bias for the consumer.

Everyone is getting fed what they want to eat, believing that because they agree with it, it must be correct.

People seem to think that I am negative on AI, but I am not, it is going to play a vital role in the future. However, whether it is going to be a tool that improves us, or one that enslaves us through our own laziness and complacency, is the question. In a world of decentralised, transparent and trustworthy data, AI can be an amazing help. But in an opaque box, fed disinformation from a centralized source - do you really want it to become your god?

Praying is useless.

Taraz
[ Gen1: Hive ]



0
0
0.000
51 comments
avatar

I've been saying for a while now that I don't really feel like ChatGPT is actual AI. At least not the machine learning piece of it. It isn't learning anything, it is just regurgitating. Which I guess opens up a new (or old depending how you look at it) conversation about the education system and if just memorizing rote facts is actually learning. I think it is impressive for sure, but nowhere close to where a lot of people think we are in terms of AI. I'd posit that the computers that predict the weather are closer to AI than ChatGPT. They are at least learning patterns and applying that to future models/forecasts.

0
0
0.000
avatar

Yep. It is a regurgitator, but because most people do the same at a lower level, they are impressed by what it spews out. Most do not think for themselves.

0
0
0.000
avatar

I've read quite a bit of content from it and I do think it does a good job of summarizing concepts, but a lot of the stuff is just too generic to be useful. Some of these AI price predictions for future crypto prices is a perfect example. It basically does everything it can to avoid the question.

0
0
0.000
avatar

"I am not giving financial advice"

0
0
0.000
avatar

I have a hard time accepting any god let alone attributing that role to an artificial intelligence heh heh!

We have already talked about it, I believe that AI is a tool and therefore in itself it is neither good nor bad in an intrinsic way, it always depends on who uses it and how they do it.

That said, as far as Hive and content creation is concerned, I will continue to reject it with all my might; I just wrote a weekend engagement post where I really talk about my passion for writing.

Having things written down doesn't convey any emotion and leaves me with nothing about whoever wrote that text; I prefer a thousand times a text written badly orthographically but which conveys emotions and feelings.
On the other hand, there are many posts on hive that I don't consider informative, especially the technical ones; if I want to train and inform myself, there are many manuals and I don't think a 500-word post is enough, moreover there are also specialized sites on each topic on which to be able to deepen much better than a post written quickly to recover some prizes.

In the same way, there are beautiful long and exhaustive informative posts that are also very useful from a technical point of view, however, the commitment and time dedicated to it show and have nothing to do with artificial intelligence or with the skimpy posts of which I wrote above.

I may be "vintage", "differently young" but I still prefer to write my posts and articles myself and I think I will continue on this path.

0
0
0.000
avatar

I have a hard time accepting any god let alone attributing that role to an artificial intelligence heh heh!

I'd more easily accept the AI one! :D

I agree with you on the "technical" posts too - there are lots of them that don't actually tell anything - just more spun content from the internet. What I do like are the "DIY" posts where people actually do things and, show parts of their life and personality through the content. But it is about the person, not the content in these cases.

0
0
0.000
avatar

I would more easily accept the AI ​​one! :D

Heh heh me too.

Yes, I agree that a lot has to do with people more than with the content, the fact remains that certain contents, as far as I'm concerned, are much more captivating and usable than anything else and are usually always written by the same authors.

I was also reflecting and if we think of the entire ecosystem there aren't even a lot of them, by eye every day I think that no more than 20% are really nice posts to read... all in all it's not even a bad percentage.

0
0
0.000
avatar

I don't where I stand with this AI debate.
I think these new suites of products are "tools". They can be used to generate facts, structures, or design. But, they may not be right tools to produce opinions, rationales and qualitative analysis. I also think these tools should credit the articles and products they rely on to produce the content so that users know where that information is coming from. I found bing is doing that with their chat functionality.

0
0
0.000
avatar

I also think these tools should credit the articles and products they rely on to produce the content so that users know where that information is coming from.

Yet they don't - ever wondered why? It is because likely, a lot of the content they use is not overly great quality, and so generic it is useless for all.

0
0
0.000
avatar

I tried Chat GPT to see how they would handle research articles. The paragraph noted one article when I specifically asked on the chat function. I could not find that article when I searched manually. Not sure where that article was pulled from.

And, the paragraph that came had errors throughout. I am sure the machine learning can improve the quality over time as the more we use the product the cleverer it becomes.

0
0
0.000
avatar

You've got an interesting take on it, if someone were to write about your company, don't you think they would also use the information already available? On certain topics, there is only so much information.

If I asked it to tell me about sloths, it would give me facts and information that is widely available, is it plagiarising, or is it simply giving the relevant information? As you say, it picks sentences from your original content, about the subject you've asked. It's trying to be as accurate as possible.

The main issue on Hive, as you say, is people will exploit it to get quick rewards, this is of course something we should be against if we value our own creations and the rewards we rightly deserve. Just like anything where money is involved, unscrupulous actors will take advantage, Ai most definitely will become a larger part of all our lives as we move forward. We Need to learn how to embrace it to benefit us and the quality of content we can produce, without taking away from the Human aspect.

Ask it to create a story/poem/ something creative, it'll give you something different. As we continue to pump information into the world, eventually, everything will have been said in a unique way it can be and everything will become plagiarism to an extent.

0
0
0.000
avatar

is it plagiarising, or is it simply giving the relevant information?

Giving relevant information, like a search engine. But if you take that and create a sloth post on Hive for reward, it is plagiarising. Perhaps if the AI was charging for the information it was giving, it would get into trouble - unless they have a system to micropay the original authors of course :)

eventually, everything will have been said in a unique way it can be and everything will become plagiarism to an extent.

And, the more powerful the AI, the faster it does this and it could be that everything is discovered in seconds. Then what do we do with ourselves as humans? :D

0
0
0.000
avatar

Giving relevant information, like a search engine. But if you take that and create a sloth post on Hive for reward, it is plagiarising.

Sounds fair, no need to worry about us sloths, we may be lazy but we like to try buzz our own stuff! Do love the ease of making images though, and AI has been useful for ideas! Plenty of off chain ideas to consider too!

When AI takes over, we'll all be left to sloth it out in our 15-minute cities, with everything we need in walking distance! Chill in your tree and watch the world go by.

0
0
0.000
avatar

When AI takes over, we'll all be left to sloth it out in our 15-minute cities, with everything we need in walking distance! Chill in your tree and watch the world go by.

Trapped in your own tree, with the ground patrolled by tigers :D

0
0
0.000
avatar

I was reading a post last night in which one of the commenters was going crazy defending the use of AI. I had a look at her blog, whatdya know she has been pushing AI shite for weeks. Lol

0
0
0.000
avatar

No surprise really, is it? People who can't or have struggled to build a following (generally because they don't produce much worth following), then justify taking this path as the rewards to actually create aren't enough - which is ultimately because they aren't cutout to be a creator anyway.

0
0
0.000
avatar

Yes. It all comes down to entitlement. I should get what I want however I want because I am special. Blah blah. Twits

0
0
0.000
avatar

Notice how the "however I want" is never by working hard?

0
0
0.000
avatar

Slightly different but it makes me happy is the Wordle Word Solver.

You feed in any correct, correct but misplaced and incorrect letters that you have, and it produces the list of words it might be. I think there's only been one occasion when I had one possible word the solution could be (kiosk - that is a pretty unusual combination of letters). Every other time there is a list of words and - the bit I enjoy - I get to choose which I think is the most likely solution. And sometimes it takes me all the remaining four goes to get the right word, even with AI help 😂.

The few AI posts I've read are well written but are immediately suspect (apart from perfect grammmar, spelling and layout) because they have very little depth. While I didn't understand immediately that what I was reading was AI generated, it was off in some way. If you know anything about the topic you're reading about, the text immediately generates lots of questions because, in some way, it's skating or skimming over the surface. And if you read a series of AI posts, one after another, it's curiously unsatisfying. It's a kind of equivalent of ultra-processed food for the brain. Given the damage that ultra-processed "food" does to us, I wonder whether there will be similar consequences for cognition, imagination etc through using AI

(I agree with all your points).

0
0
0.000
avatar

I have never played Wordle, but isn't it cheating? I don't get it! :D

The few AI posts I've read are well written but are immediately suspect (apart from perfect grammmar, spelling and layout) because they have very little depth.

Yes - because it is all general information, mneaning it is largely useless and available on Google anyway, which is also another reason niot to pay anyone to produce it.

While I didn't understand immediately that what I was reading was AI generated, it was off in some way

The "uncanny valley" of text :D

Given the damage that ultra-processed "food" does to us, I wonder whether there will be similar consequences for cognition, imagination etc through using AI

I believe it will absolutely be the case. We are feeding ourselves mental crap, feeling like we know something useful.

0
0
0.000
avatar

Cheating? 😎

0
0
0.000
avatar

Like I said, I haven't played the game - but are you using a helper of some sort?

0
0
0.000
avatar

Yes, it's a form of AI, or an algorithm, that's why I was telling you.

The game is to guess a five letter word in six goes, with each try, the game will tell you if you have correctly placed a letter in the word, or the letter is in the word but in the wrong place, or the letter is not in the word; so with each guess, you start to get clues about the final word.

It is quite amazing how difficult it is sometimes to guess any words that might have the letters in that you do know. I have ruminated for hours over the next guess. The helper, the AI bit, presents you with a list of words with your already known letters in them.

It would be a form of cheating if you were in a competition or posting it on Hive.

0
0
0.000
avatar

ah okay. sounds like an interesting game, though one I would have been great at "back in the day" but would suck hard at now. Might give it a go, just to make me depressed. I used to love word games....

It is a bit like hangman by the sounds.

0
0
0.000
avatar

I struggle sometimes when I am tired.

Yes, it is a bit like hangman, that's a good comparison.

I've never commented about your stroke, it always felt too intrusive (or the danger of being facile about someone else's experience), even though you have been very open about it. Several people around me have had strokes, including my previous partner. It was very tough, but I read a huge amount about it, and about recovery and how long it takes, and although my partner didn't recover, I was heartened by the possibilities of recovery.

0
0
0.000
avatar

AI might be useful for cooking. I wrote ,,herring with mushrooms recipe''. It responded with

Herring and mushrooms are a tasty combination. Begin by heating oil in a skillet and sautéing cleaned mushrooms on medium-high heat. Cook until softened. Crisp up herring fillets on both sides in the same skillet. Serve the dish topped with the cooked mushrooms. Delicious!

,,herring with orange recipe''

This tangy, flavorful dish is sure to please. Start by stirring-frying herring fillets in a pan with a bit of oil for about four minutes. Add freshly sliced oranges and cook until the oranges are just beginning to soften. Sprinkle with herbs like parsley, oregano, or thyme, and season with salt and pepper to taste. Serve with a side of potatoes or a bright green salad. Enjoy!

Seems a bit more to the point and shorter than google search. Well that might be a good or bad thing depending on opinion.

0
0
0.000
avatar

Yes. So as a search engine, it might be okay - as a creator, not so okay. Can it make a recipe that tastes great, but has never been made before?

0
0
0.000
avatar

I guess people don't really know that ChatGPT actually collects information from the WEB and then types it down in its own words and style. I guess some people really think that the bot itself is the one writing and compiling words from its own thoughts lol.

So far I've used ChatGPT for one or two tasks, and one time it really did help. Me and two of my friends were planning a new business idea for quite a while now, and we needed a name for the establishment. So we hit up ChatGPT, and whaddayaknow, we actually ended up liking some of the names. We even shortlisted some of them.

For now that's the only scenario where I actually thought it was useful. Other than that ChatGPT still feels quite basic to me, or maybe I'm just not capable of pushing it to the limits?

AI generated images though, I say there's some creativity in there.

0
0
0.000
avatar

I don't know what people assume with it - but most thinks it is smarter than it is, because it is smarter than them :)

That is an interesting use for it. I also think it might be useful for various things, especially if sorting through and making sense of large amounts of content. Later, it will be much better, and more people will be out of a job.

or maybe I'm just not capable of pushing it to the limits?

I know one highly skilled user that is using it to learn languages and such, but he is essentially coding it to do various tasks for him automatically. It is pretty cool and will improve, but it has its limits for content creation.

AI generated images though, I say there's some creativity in there.

Some, but not much. It might be on the spectrum of art, but definitely not high art, where practical skills are needed. If a person uses it to design a building, then gets that building made, it might be interesting :)

0
0
0.000
avatar

because it is smarter than them

Ain't that a fact. 😂

and more people will be out of a job.

A lot of such software and bots are already taking over, it's just a matter of time now.

but he is essentially coding it to do various tasks for him automatically.

In terms of creativity and actual usefulness the possibilities are endless, but instead of putting it to good use, most people are just abusing it.

If a person uses it to design a building, then gets that building made, it might be interesting.

Oh, that's surely something I'd like to see.

0
0
0.000
avatar

I haven't checked ChatGPT yet, but it has been trend recently. As far as I have understood from your post, it is kinda search engine.

0
0
0.000
avatar

Kind of a search engine, but people are cutting and pasting the results as if they are a post :)

0
0
0.000
avatar

I think the business of Hivewatchers is not difficult then.

0
0
0.000
avatar

Text generated from AI that is designed to learn patterns and structures of language from given data and then uses that to create responses to prompts is definitely not plagiarism.

That said, for some reason, as far as I know even developers don't understand, ChatGPT sometimes takes whole paragraphs and mixes them into generated content. In which case that is plagiarism. This is a bug, not a feature.

The important difference here is that ChatGPT can and is creating new content no matter how low effort it is. It can "invent" words and assign them meaning that aren't even a part of its data set which, at least to me, proves this point effortlessly.

We asked it to write a sales pitch... While it was able to write in natural language well, it was also taking content almost directly from available marketing material and content that both of us know quite well.

I'm really not sure what you expected here. I get the sense from these arguments that people think they aren't "copying" material they learned from, that their sentences aren't structured similarly (in some cases exactly) to people they most frequently talk to or follow their work. Just by saying that you know that particular material quite well, it means that it has a huge influence on the way you go about it too.

Just like a game, each type of work has its own set of rules and objectives. In order to succeed, it is important to learn these rules and understand what is expected of you. Playing the game of work right involves not only meeting expectations, but also striving to excel and exceed them. This can require a combination of skill, effort, and strategy. However, playing the game of work wrong can have negative consequences, such as poor performance. This applies to sales pitches too, of course.

What I like about the usage of AI, is it is raising questions of what is valuable, something that has to happen going forward.

From all AI talk, this is the most important question people don't really delve into. As we already know. When some invention automates or greatly boosts the productivity of a specific task, and if we have an abundance of material to do the task that task becomes less valuable.

Fully hand-off AI-generated texts will very quickly become the floor of writing skills. Any writing that isn't more interesting, creative, engaging, etc. than that kind of writing won't be worth much.

However, whether it is going to be a tool that improves us, or one that enslaves us through our own laziness and complacency, is the question.

To me, the answer is plainly obvious. My answer is no. Because technology itself never made us lazy. The abundance of food and shelter, and safety in general makes us lazy. Some people when they come from work just want to lay on the couch the rest of the day and watch TV. Would they be able to do it without the things listed above? I don't think so.

0
0
0.000
avatar

Text generated from AI that is designed to learn patterns and structures of language from given data and then uses that to create responses to prompts is definitely not plagiarism.

Posting that text into Hive for rewards though, is plagiarism, right?

Just by saying that you know that particular material quite well, it means that it has a huge influence on the way you go about it too.

I wrote some of the lines it used ;D

Fully hand-off AI-generated texts will very quickly become the floor of writing skills. Any writing that isn't more interesting, creative, engaging, etc. than that kind of writing won't be worth much.

And this is exactly the point. I can add and subtract quite well, I do not consider myself a mathematician. Using AI to generate my texts, wouldn't make me a content creator. As I have written, proof of person and trusted sources, as well as ownership, are the future. If you don't own your skill, it isn't yours.

Some people when they come from work just want to lay on the couch the rest of the day and watch TV.

Some people don't have jobs to come from - they just watch TV.

0
0
0.000
avatar
(Edited)

Posting that text into Hive for rewards though, is plagiarism, right?

Absolutely no.

I wrote some of the lines it used ;D

So what? You think you are the original source of your thoughts, think again? Most people would disagree.

Some people don't have jobs to come from - they just watch TV.

These people will use the bottom line of the tech, the same as people use a calculator to calculate 212x13.

Your points are as weak as they can be. I'm not saying that to belittle you or anything it just seems that you didn't give it as much thought as you could. It just sounds like your emotions cloud your judgment here. Like your skill aren't as unique as you would want. The reality is that most of the writing skills are going to get replaced as much as people doing manual calculations were replaced when the first electronic calculators were replacing humans (we are talking about hundreds of thousands here) and you don't like it. But that doesn't matter. It is happening, and we need to prepare instead of arguing. These half baked arguments are just waste of our time.

0
0
0.000
avatar

I just want to add.

Something you don't want to get rewarded doesn't have to be plagiarism. If you don't see the effort you shouldn't vote or should even downvote, right? Although we can all see posts (not too many but still) on the hot page that are low effort yet these rules aren't applied to them. So I doubt AI content will get a different treatment depending on the user posting it.

0
0
0.000
avatar

Thanks you for this interesting analysis :)

Completely agree with your points and reflections here. I find that it’s a prodigious tool but indeed not to produce a true, honest and even meaningful opinion! Rather use it for some research and list that would too time taking !

I also thought about building my own, but even if it works, it will be completely irrelevant to right “for me”...

0
0
0.000
avatar

And the lists it creates are generic lists, not opinionated lists, so what is the value in paying for them. It is like a one-size-fits-all piece of clothing - it doesn't mean it looks good :D

0
0
0.000
avatar

it doesn't mean it looks good :D

😂 Indeed !

I'm rather using it for gathering More data and then I do my sorting... For example, I have him make lists of wild edible plants or tables of photographic data (apertures, speeds, etc). But in the end I only use it as a search engine that is improved in some aspects... and almost deficient in others !

0
0
0.000
avatar

I think detection tools are desperately needed. Because nobody wants to have machine generated content. Because unless we are getting real life manquin robots who give us company. I don't see the value of anything on that front. AI is going to help introverts a lot. They will fill the gap of dead people in family or even don't let anyone be alone in life. So that is one positive but AI taking the job is where I am against to be honest.

0
0
0.000
avatar

There are detection tools, but they themselves have a wide range of outcomes. After testing a few well-known creators here, there are some "discrepancies" in what they are putting out.

0
0
0.000
avatar

After testing a few well-known creators here, there are some "discrepancies" in what they are putting out.

'Well known creators'.., did you find something?

0
0
0.000
avatar

Nothing reliable enough to mention due to the tools available. But, should there really be any questions?

0
0
0.000
avatar

The detectors claim to be 99% accurate, are you skeptical of their boasts?

0
0
0.000
avatar

Definitely skeptical. Scroll up a little in the Discord chat and you can see.

0
0
0.000
avatar

Using AI to generate posts is not only plagiarism and cheating, it's also plain DUMB. Why would you do that? What's the purpose? To make money? There are so many other smarter ways to make money out there... I just don't get it..

0
0
0.000
avatar

I don't know much about it but it does feed on user data so it will spin for sure and now big tech giant like google are in race to make ai

0
0
0.000
avatar

Very interesting that the chatbot is taking existing content and spinning it.

I can see content creators starting to block this using the robots.txt part of their website, where they can block named bots from crawling their content. Presumably if you just block the Bing crawler, you block their chatbot too (of course that means your site won't get listed in the Bing search results, but if you don't get much traffic from Bing, it doesn't matter).

0
0
0.000
avatar

I has to deal with this very issue this week. As content Manager, I have a single long-form writer to provide our paid content and I edit it. However, he was away on vacation this past week leaving me without a writer to handle any new assignments and of course, the Tuesday I received some new assignments. With permission from my director, I used ChatGPT with 5 different writing prompts to come up with long-fork copy similar to what I would get from my writer.

Then I aggregated the similar content and went about editing the article to get the tone and phrasing more in line with the way we publish. I wa unfamiliar with the topic source, so I had no clue about plagiarism. But I felt like I did a good enough job editing the content so that it felt much more like something of my own work than an article generated by AI. To me, it just felt like a much more elaborate search engine, that summarized the answer in complete sentences and thoughts instead of passing me an encyclopedia e try and making me do the processing myself. But at the very least I've finally opened Pandora's Box (at work) and there is no way to keep it closed...

0
0
0.000