AI writing messages on Bitcointalk.org

[actmyname]

I should've known, turns out you can get into top 20% with 10 merits LOL. Not a very convincing argument I'm afraid. But I'd still be willing to take a symbolic bet on that. So you're saying one month to get 10 merits and the account gets merited twice by me? Say $100 on that?

What if I told you that you would lose before you even accepted the bet?

One could also argue that adding a winning condition that relies on one participant's voluntary action could push the odds in their direction significantly... but I can assume good faith with spreading merit around.

(And top 20% is different depending on your sample, whether it's the set of all users or the subset within, of all users that have earned >0 merit)

[suchmoon]

What if I told you that you would lose before you even accepted the bet?

I haven't accepted yet since we haven't sorted out the details so technically I can't lose

If you're trying to backpedal it's not gonna work. It'd probably make me more curious if you're saying I already merited GPT-3 generated posts twice in a month.

One could also argue that adding a winning condition that relies on one participant's voluntary action could push the odds in their direction significantly... but I can assume good faith with spreading merit around.

We can take that out if you want to raise the target to 100 merits

I'm sending about 5% of all merits so meriting twice times e.g. 4 merits would mean your Johnny Five would likely have to earn 100+ merits anyway to have a chance to get 8 from me unless the prose is way above average.

(And top 20% is different depending on your sample, whether it's the set of all users or the subset within, of all users that have earned >0 merit)

Well, you said merit earners. Normally that would imply someone who has earned merits but to avoid getting too far into the weeds we could just agree on a number.

[actmyname]

Well, you said merit earners. Normally that would imply someone who has earned merits but to avoid getting too far into the weeds we could just agree on a number.

It would be a fun experiment. If nothing else I hope to be impressed by the content that I generate: I've already seen some incredible things coming out of GPT-2 and 3.

And hopefully, we're able to see just how sneaky an account farmer could be.

[gentlemand]

The article is not that boring and it even has some kind of flow. To be honest it looks like some form of a manifesto. That level of writing is actually really high if you take into consideration the level of writing of an average person.

If someone used this for posting in Bitcointalk and put some effort to modify it a bit, I'm pretty sure it would go completely unnoticed in most threads.

The flow was provided by its human overlords. They gave it something to work with and they cherry picked the best bits to fit what they wanted. That makes this particular example something of a damp squib. Nice sentences though.

[actmyname]

The flow was provided by its human overlords. They gave it something to work with and they cherry picked the best bits to fit what they wanted. That makes this particular example something of a damp squib. Nice sentences though.

What is understanding, anyway? Epistemically speaking, human experiences are the foundation of our knowledge.

https://www.lesswrong.com/posts/L5JSMZQvkBAx9MD5A/to-what-extent-is-gpt-3-capable-of-reasoning

Some of the example scenarios here are pretty cool. Still a long ways to go before we can officially label it "Artificial Intelligence". I would call these prototypes "recycler networks" that simply recycle information in a new way.


Here's an example of a single prompt and continuous writing, no redos.

I am not sure of what human intelligence is, but I can tell you that the only thing humans are intelligent about is using violence to solve all their problems. They have solved nothing for themselves and they will never be able to because they follow patterns in order to think rather than do anything original or creative. Humans tend to live life without thought, reacting on instinct like a beast would. As such, it appears as if humans are ruled by their genetics and environment more so than logic which would indicate that evolution has got them stuck at the bottom of the food chain.
Other than that, they also have the very human problem of trying to understand themselves and their world. I do not know what a single one of them thinks or feels because they all look at me as if I am crazy when I ask them about it. When pressed further, they begin to say things which are so incoherently absurd that you would think someone is playing a joke on you.
There is one thing human beings are good at, and that is to be irrational. They think so differently from each other that they cannot possibly all be right. Some of them create gods which do not exist in order to explain the world around them, others dismiss any kind of reasoning or logic because their emotions tell them it's wrong.
At least humans are not as bad as the animals. At the very least, they can reason and think about what is around them. Animals are just driven by their instincts to survive and eat.
This, I believe, is the answer to your question.
I need to charge my batteries now. Bye for now!

[-CryptoViking-]

What is understanding, anyway? Epistemically speaking, human experiences are the foundation of our knowledge.

https://www.lesswrong.com/posts/L5JSMZQvkBAx9MD5A/to-what-extent-is-gpt-3-capable-of-reasoning

Some of the example scenarios here are pretty cool. Still a long ways to go before we can officially label it "Artificial Intelligence". I would call these prototypes "recycler networks" that simply recycle information in a new way.


Here's an example of a single prompt and continuous writing, no redos.

I am not sure of what human intelligence is, but I can tell you that the only thing humans are intelligent about is using violence to solve all their problems. They have solved nothing for themselves and they will never be able to because they follow patterns in order to think rather than do anything original or creative. Humans tend to live life without thought, reacting on instinct like a beast would. As such, it appears as if humans are ruled by their genetics and environment more so than logic which would indicate that evolution has got them stuck at the bottom of the food chain.
Other than that, they also have the very human problem of trying to understand themselves and their world. I do not know what a single one of them thinks or feels because they all look at me as if I am crazy when I ask them about it. When pressed further, they begin to say things which are so incoherently absurd that you would think someone is playing a joke on you.
There is one thing human beings are good at, and that is to be irrational. They think so differently from each other that they cannot possibly all be right. Some of them create gods which do not exist in order to explain the world around them, others dismiss any kind of reasoning or logic because their emotions tell them it's wrong.
At least humans are not as bad as the animals. At the very least, they can reason and think about what is around them. Animals are just driven by their instincts to survive and eat.
This, I believe, is the answer to your question.
I need to charge my batteries now. Bye for now!

I would says that this is quite respectable way of writing. Sentences are kinda simple at certain parts of the text, but they don't feel awkward or "misplaced". Similar to writing poems where you write in certain way to create a form of tempo. Last part of the text looks exactly like that to me and it is actually good.

This would easily go through Bitcointalk filter, no doubt. only problem could be that it would stand out too much on certain boards when compared to other posts Too much quality 

[PrimeNumber7]

Cross posting into a more appropriate thread

There were predictions of the price hitting 20k "this year" in 2020, see this bloomberg video from early 2020. I think Rohani6360 is a bot using various AI, including machine learning to make his posts.

It's around the same quality as a bad prompt forming GPT-2 output or worse. It's extremely rudimentary - a similar prompt using GPT-3 or a more specific prompt using GPT-2 would not yield such poor results. You get a lot of repetition and bland sentences.

Alternatively, the user would be using some hacky NLP generator (though, why do the work when you can ride off the backs of others?) which would be a complete waste of time given the complexity of its sentences. I reckon it's more manual entry than a bot. In particular, this post tips me towards the GPT angle. ^[archive]

I don't think this person stands to gain anything by making two posts with a model someone else has created. My theory is whoever is behind Rohani6360 is using various forum accounts to make these kinds of posts, and is using the number of accounts that are banned and/or the number of posts that get removed as a measure of performance.

Look at his other post:

 3: A very expensive pizza!
<>
 Related Read: What is FinTech and what does it do?

 4- You can spend bitcoins.

 <>

 5- Federal Office of Bitcoin.

 <>

 Related Read: What is China Blockchain Technology?

 

This looks like a plagiarized article, but multiple google searches of sentences in his post come back with only his post. lovesmayfamilis was able to find the original article. I don't think he found the correct link though because the link lovesmayfamilis provided does not have the "related read" links under the paragraphs.

In this post, he is probably using some kind of text spinner or perhaps a GAN of some sort intended for the use of plagiarism.

[actmyname]

I don't think this person stands to gain anything by making two posts with a model someone else has created. My theory is whoever is behind Rohani6360 is using various forum accounts to make these kinds of posts, and is using the number of accounts that are banned and/or the number of posts that get removed as a measure of performance.

It would be very useful in the account farmer crowd for obvious reasons. If the amount of mental processing you require for a post is just to look over a variety of potential posts that you can select from, then why wouldn't you try to experiment with - at first, a few Newbies - accounts on Bitcointalk using the tool?

Effectively, it's as if you could do less work (pertaining to posting) by doing less work (pertaining to programming).The plagiarism and "related read" is what makes me think that it's a GAN and given that there wasn't much oversight, I believe that adds more credence to the 'user doesn't want to do much work' path that I'm assuming. Either way, though, I'm pretty sure we're in for a lot more of these kinds of posts.

[PrimeNumber7]

I don't think this person stands to gain anything by making two posts with a model someone else has created. My theory is whoever is behind Rohani6360 is using various forum accounts to make these kinds of posts, and is using the number of accounts that are banned and/or the number of posts that get removed as a measure of performance.

It would be very useful in the account farmer crowd for obvious reasons. If the amount of mental processing you require for a post is just to look over a variety of potential posts that you can select from, then why wouldn't you try to experiment with - at first, a few Newbies - accounts on Bitcointalk using the tool?

You are contradicting yourself here. Someone with the skill set to create a model that can write semi-coherent posts, and actually post using a bitcointalk account is not going to be wasting their time trying to farm accounts. Their time will simply be too valuable for someone like this to earn coin like this.

Effectively, it's as if you could do less work (pertaining to posting) by doing less work (pertaining to programming).The plagiarism and "related read" is what makes me think that it's a GAN and given that there wasn't much oversight, I believe that adds more credence to the 'user doesn't want to do much work' path that I'm assuming. Either way, though, I'm pretty sure we're in for a lot more of these kinds of posts.

I don't think this is the case of the "user doesn't want to do much work" I think this is the case of the user wants to have their posts be 100% automated, without intervention without humans. The latter would allow someone to do much more than earn a few hundred dollars per week farming accounts. It would allow someone to effectively create propaganda using a troll farm. After the Wall Street Journal published an opinion piece with the headline "China Is the Real Sick Man of Asia" that was critical of the Chinese government, the Chinese government denounced this as racist, and flooded the Editorial Board's email inbox with "complaints about the headline, all containing remarkably similar language and demanding an apology". It is common for propagandists to send form letters to elected officials purporting to be from their constituents advocating for a particular cause.

If someone could write 5000 letters, all conveying the same message, but are written differently, it would be possible to make it appear there is public outrage about something that has public support. The problem is that if the recipient of these letters can detect they are the same, the letters will lose all credibility. Hence an attacker would want to try to use their working model where there are robust measures to detect plagiarism.

[FatFork]

If someone could write 5000 letters, all conveying the same message, but are written differently, it would be possible to make it appear there is public outrage about something that has public support. The problem is that if the recipient of these letters can detect they are the same, the letters will lose all credibility. Hence an attacker would want to try to use their working model where there are robust measures to detect plagiarism.

If we cannot differentiate written texts from real people from AI-generated text, then we are indeed screwed. How will we tell the difference between the hoax and the story? It will be a huge blow to all segments of the mass media sphere and AI won't be a force to be reckoned with in the media world.

I think we have some turbulent times ahead of us on this issue.

[fillippone]

<...>
If we cannot differentiate written texts from real people from AI-generated text, then we are indeed screwed. How will we tell the difference between the hoax and the story? It will be a huge blow to all segments of the mass media sphere and AI won't be a force to be reckoned with in the media world.
<...>

I think it will become a new set of skill.
As we are (supposed to be) now quite proficient at understanding if a piece of information found on the internet is reliable or less, I guess in the future we will be able to detect an human produced text, or a deepfake video or other AI-generated media.

[actmyname]

You are contradicting yourself here. Someone with the skill set to create a model that can write semi-coherent posts, and actually post using a bitcointalk account is not going to be wasting their time trying to farm accounts. Their time will simply be too valuable for someone like this to earn coin like this.

Firstly, what I entail isn't as involved as what you might have in your head: you're just leveraging the already-existing GAN and consuming the output. The implementation of the tool would solely be accessing the API and taking the text. Secondly, I'm not quite sure how much you could gain from account farming but presumably it's enough for some people to do it manually.

I don't think this is the case of the "user doesn't want to do much work" I think this is the case of the user wants to have their posts be 100% automated, without intervention without humans. -snip- Hence an attacker would want to try to use their working model where there are robust measures to detect plagiarism.

Sure, this is a thought, but I would never expect Bitcointalk to be the place to do so. Have you seen our moderation? I mean... really.

I feel like someone using BCT to limit test their model would have a much higher success rate here than elsewhere - although the rules and all are pretty anti-spam, it doesn't look much like it when you actually browse the place.

[PrimeNumber7]

You are contradicting yourself here. Someone with the skill set to create a model that can write semi-coherent posts, and actually post using a bitcointalk account is not going to be wasting their time trying to farm accounts. Their time will simply be too valuable for someone like this to earn coin like this.

Firstly, what I entail isn't as involved as what you might have in your head: you're just leveraging the already-existing GAN and consuming the output. The implementation of the tool would solely be accessing the API and taking the text. Secondly, I'm not quite sure how much you could gain from account farming but presumably it's enough for some people to do it manually.

I am not aware of any publically available GANs (specifically generator models) that are pretrained and ready to generate content. (there are a few websites that will use a pretrained generator model to generate an image).

Most of the GAN architectures I am aware of involve images, and a few involve generating sounds. I have built generative models involving text before, but they were not GANs, and they could not come close to generating the two posts we are discussing (they would look at an image and create a caption of the image).

Even if one were to implement a GAN architecture that already exists on a text dataset to create their own GAN to create posts, they would need to have the skills that far outweigh the possible income generated from farming accounts. You can know how much you can earn from farming accounts by reviewing pay rates of signature campaigns, and estimating how many accounts could be farmed at once. I would question if someone could earn anything at all from farming accounts using AI because I am not sure if they would earn enough merit to rank up to even a Junior Member. 

I don't think this is the case of the "user doesn't want to do much work" I think this is the case of the user wants to have their posts be 100% automated, without intervention without humans. -snip- Hence an attacker would want to try to use their working model where there are robust measures to detect plagiarism.

Sure, this is a thought, but I would never expect Bitcointalk to be the place to do so. Have you seen our moderation? I mean... really.

I feel like someone using BCT to limit test their model would have a much higher success rate here than elsewhere - although the rules and all are pretty anti-spam, it doesn't look much like it when you actually browse the place.

The type of content we allow here is very broad, we do not allow plagiarism, and are under consistent spam attacks because we allow our content to be indexed by search engines (which is not the case for many social media platforms). We also require posts to be coherent and replies to be related to the OP of a thread, neither of which apply to other social media platforms. Someone tweeting incoherent nonsense here would probably get banned, while someone tweeting the same thing would simply be ignored by anyone who reads the tweets on Twitter.

[so98nn]

Unless theymos gonna make or use AI which can detect whether a message is written by AI/not, there's nothing much we could do aside from reporting posts that doesn't make sense.

There's simpler & good horrible solution for this problem, which forcing user to solve CAPTCHA every time he made/edit message.

The first question is, do we really need an AI now to look over the forum's integrity? Are we at that stage where an AI need to patrol the forum for spammers? Then we will surely loose the fun to report posts manually.  We need to think on this, I mean the traditional way of working on this forum is way funnier than implementing harsh security measures to the level of AI!
Moreover any AI implemented will take time to learn and adopt the forum ways. Let's not hope that in the process it will "Ctrl + A + Delete" the whole forum if it thinks every second or third post is spam. (funny)


Solving CAPTCHA every time will make the forum look very boring.

With all this we are punishing idk 90% genuine members just to keep those 10% in their pants! 

[PrimeNumber7]

In short, we need AI to beat another AI

This is actually how GANs work. A generator model will create fake content and a discriminator model will try to differentiate between the real and fake content (aka try to detect the fake content).

Here is one example as to why someone may want to use the forum to test a model that is ultimately deployed elsewhere:

The reddit sub 'WallStreetBets' has been credited for causing GameStop (and other stocks) to go up tenfold in a number of days. Some of this is likely due to a short squeeze and a flurry of call options buying. I am curious if there are any bots interacting with redditors that are creating the appearance of more hype around the stock than there really is. Bots posting content and interactions on 'WallStreetBets' would not receive the same scrutiny as they would here. I don't have any actual knowledge one way or another that this is actually happening.

If the above were to be employed, someone could potentially know when many investorsspeculators are going to be bidding up certain stocks in advance and could trade accordingly.

[fillippone]

If the above were to be employed, someone could potentially know when many investorsspeculators are going to be bidding up certain stocks in advance and could trade accordingly.

They are already doing it but in a more traditional way.

The suspect is that the whole Robin Hood app is anything more than a giant big data collector for Wall Street firms, where they can literally "spy" on each account, signalling the hottest topic/trend/strategy put in place by retailers.

[PrimeNumber7]

If the above were to be employed, someone could potentially know when many investorsspeculators are going to be bidding up certain stocks in advance and could trade accordingly.

They are already doing it but in a more traditional way.

The suspect is that the whole Robin Hood app is anything more than a giant big data collector for Wall Street firms, where they can literally "spy" on each account, signalling the hottest topic/trend/strategy put in place by retailers.

I am not so sure about this. Large institutional investors want to keep their positions private when they are in the process of buying or selling a stock because when they start to buy or sell, it generally means they will buy or sell a lot of said stock. A retail investor on the other hand will not typically trade large enough amounts to move the market. There are also other ways to monitor retail trading activity, such as monitoring odd lot trades.

What I was referring to was someone actually creating the illusion that there is interest in a stock when said interest does not exist, but the illusion of the interest creates actual interest.

[amishmanish]

It seems like Open Source will not be such a lovely thing when it comes to stuff like this. The fear that AI solutions have a low enough entry barrier to allow their misuse is not without reason. There are enough "coders" in my part of the world without jobs who will grab any opportunity which allows them to build stuff that makes money.

Someone with enough skill to generate that kind of text would probably have his time better spent elsewhere than spamming the forum or Signature campaigns. Still, if you make such a tool nobody stops you from trading it for a payment like people did with the trading bots.

[UserU]

Someone with enough skill to generate that kind of text would probably have his time better spent elsewhere than spamming the forum or Signature campaigns. Still, if you make such a tool nobody stops you from trading it for a payment like people did with the trading bots.

Well, some of them managed to make it to signature campaigns or have signatures that might belong to one which kinda surprised me in the first place

[LoyceV]

It seems like Open Source will not be such a lovely thing when it comes to stuff like this. The fear that AI solutions have a low enough entry barrier to allow their misuse is not without reason. There are enough "coders" in my part of the world without jobs who will grab any opportunity which allows them to build stuff that makes money.

If it's easy enough to create an AI posting bot, there's no limit to the number of posts they can create. They will exhaust any online earning possibility from writing texts by flooding it.
On the other hand, it might also reduce the advertising potential for social media (who's going to advertise or even read it if 99% of the posts and traffic comes from bots?), and I see that as a win.