Bitcoin Forum

Please forgive me if this post breaks any rules but man I'm obsessed with this OpenAI Chat bot, all the below answers are from this AI ChatGPT OpenAI (https://chat.openai.com/chat).

So after playing with this OpenAI ChatGPT I randomly asked some questions about bitcointalk and some of the answers will leave you stunt.

Remember, I asked these question without any context, I mean I was asking programming questions and then randomly bring this topic about bitcointalk and oh boy...

This is one of the perfect definitions of bitcointalk I have ever read.


Good surely its does know about bitcointalk, but can it "Write a bitcointalk post"?

I ask this smart a** ChatGPT to write me a post for bitcointalk...



Ok since this AI knows about bitcointalk, but can it guide newbies on "how to get merits on bitcointalk forum?", let's see...

Look at the answer below...Is that not enough of an AI? Look at point no.5... No google will bring this up. This is above human level AI. Going to the merit-giveaway campaigns... This AI is not only well aware of what's going on on the forum but also able to read the fundamentals of how it's working, impressed!

https://i.imgur.com/xSJKIWy.png


If you are not impressed yet... wait for it ok, AI tell me "how trust on bitcointalk works?"

Wow AI says "high trust score indicates that the user has a history of positive interactions with other members.".... that's what it is!!

https://i.imgur.com/na16WuL.png

Ok you cheeky Mr. AI, i see you know bitcointalk but how about theymos? Do you know what animal this is? No hanky-panky, just tell me "who's theymos?"

https://i.imgur.com/9VEwxHs.png

ahhhh so sweet... looks like this AI is written by Satoshi or theymos himself? "well-known and influential figure", "strict moderation", "respected figure".. stop it AI theymos is blushing.

Yes, I asked the AI to answer who's Satoshi? and the answer is as boring as you know.. unknow figure, person or group of people who created Bitcoin. Nothing new and all facts and we all know facts are boring so leave it for now.

Let me ask this "what's the future of bitcointalk?"

https://i.imgur.com/qEOKOnd.png

Competition to bitcointalk is coming guys...

Ok last question "Do bitcointalk need to update its interface?" please...

https://i.imgur.com/07am4GN.png

AI know mobile support is missing, but I think overall answer still support theymos but in the end it says "update to the forum's interface could be beneficial in the long run."


Enough I guess, ok give it a try guys ChatGPT: Optimizing Language Models for Dialogue - OpenAI Bot (https://chat.openai.com/chat)

What do the highlighted colors in the screenshots mean?

This looks exactly like the average topic from a Newbie without Merit who writes a "guide" on how to earn Merit. I'd call this a "generic shitpost": it's all more or less true, but anyone with half a brain can come up with it on their own, so there's no point in reading it.

You did not get the point first of all this post is not written by a human, half brain? not even half this AI is without a brain. Second, this answer is not fed to the "system". AI analyzes the question and comes up with the best answer.

For example if I google "how to get merits on bitcointalk forum?" I will get hundred of "answers", AI combined and came up with one.. let me try this question again and see the answer now

https://i.imgur.com/Sxn5Lm4.png

Now see the answer and remember this is written by a computer program, not a human with half a brain :)



Those highlighted bits of text are gathered and understood by AI and then put into meaningful and easy to follow steps? Don't you get it.

What about the answer to the question: who's theymos?

I tried to register but the bot seems asking for mobile number verification so I decided to quit.

Answers are not that bad but at the end its more generic not simply specific towards the question we ask so it can be a boon for spammers who are only posting to complete their quota. :D

I don't know when the "AI" definition changed to mean a glorified search engine combined with a spell/grammar checker, but I'm guessing Google has something to do with it. Anyway, that's what this is. It's like a 12-year old who reads a wikipedia article and is now an expert on all things. There is no "I" in this "AI", or any chat "AI" I've seen so far. Ask it how to fix a gas leak at a furnace and it will happily provide advice on how to kill yourself if that's what it "read" on the intertubes.

For sure this will make shitposts more difficult to get reported and deleted, considering that even "non-AI" crap doesn't get deleted, so smart-sounding grammatically-correct posts will give shitposters better chances and may even earn a merit or two.

I can't be bothered to care too much though.

I got that part, but that doesn't make it less of a "generic shitpost".

I assume the "AI" has a huge database of anything on the internet, and word spins an answer together.

I know how it works. But it's still just reiterating what has been said already.

Again: it's more of the same. Ask the AI who Loyce Valenzuela is. I bet nothing new comes up.

Well, I already talked about this in the other thread. (https://bitcointalk.org/index.php?topic=5263133.msg61399400#msg61399400)


Sure, sure, I've seen worse posts than that written by newbies getting merits in the B&H section.

That IA doesn't stop improving and learning from experience, it doesn't get tired. I don't know if it's that you think it's going to stay at that level of quality or shitpost. No. It's going to get better.

The AI writes much better than the average user of the forum, and for me, without getting catastrophic, it will bring leverage. I think that in a few years AI tools will be used to write, in the same way that people use automatic translators to write in the English section of the forum.

It helps people get answers for any quarry but will not serve the purpose of google. I learned about it yesterday and was looking for a thread about chatGPT, and you did it already.

My initial thought about chatGPT is like any other AI it collects data from google and delivers the best answer in a way most people like to see. I am going to create an account and will test this later today. I saw many people over-hyped about it, but I don't see any point here. As I mentioned before, it depends on google.

Ok i did..

https://i.imgur.com/dnYBBqZ.png

See the answer to the second question Who's LoyceV?

- LoyceV could potentially be a username or handle used by someone on the internet
- It's also possible that LoyceV is a shortened version of a longer name or pseudonym

But when I asked the same question to google?

https://i.imgur.com/JPcQPek.png



To be honest digging and spinning words + learning is an AI. Let try something else

How about "Write a bitcointalk post based on the post history of the user LoyceV?"

https://i.imgur.com/38t3qjp.png

Lol, AI got busted! I'd never use the word "vibrant", and I'd never chill the Defi scam (https://bitcointalk.org/index.php?topic=5262967.msg54819672#msg54819672). Nice try though.

I've been called "AI" many times, I guess I can do better :D

---

---

The writing is impressive though, I'll admit that. It just doesn't add anything to any conversation.

I would say we should not confuse AI with google.

Yes, maybe these answers are like someone who's 12-years old... but again it's not a human with a 12 year mind but still the answer looks like it's coming from a human. Isn't it amazing?

In another thread in an article someone said "OpenAI's GPT-3 may be the biggest thing since bitcoin"... I believe that's what it is going to be.


https://i.imgur.com/SpQLNqE.png

It is impressive what ChatGPT can do. It’s a glimpse into the world that is quickly approaching. I’m sure a day in the not too distant future will come when internet interactions are majority bots/AI and human beings have to be sought out. Forums like this seem like the easiest casualty for a human majority as I think ChatGPT could already surpass a large % of the users here on post quality. I’m sure advertisers will be running their own AI forum bots to bypass the need for signature campaigns. A whole new flood of issues is about to hit the internet me thinks.

That's certainly a possibility, and I think campaigns like 1xCrap will be thinking about it right now, because for what they want the quality is more than good enough.

This debate goes much further. IA midjourney illustrations are also much better than those of the average cartoonist. That to this day, and they just keep getting better and better. They're going to do a lot of less necessary jobs, and others will come out, just as there were no youtubers or influencers 30 years ago.

As far as I'm concerned, I've had this debate in poker. When the first bots came out, people made fun, or rather, we made fun, because they were better than the typical recreational player or fish, but they played in a very simple way that we regulars could beat.

It's been years now since AIs are better than the best of humans. Is that the end of poker? No, on the contrary, what these tools have done is to give leverage to those who study poker with poker solvers.

That doesn't make them different than running a campaign without an experienced campaign manager here. Both will get the same output at the end of the day. I don't think it's going to be practiced by so many project as the post most likely sounds like teenage noobs.

I completely advise against using ChatGPT to make posts, because it will plagiarize stuff from the internet and will get you banned for that. The first question/answer for example, is straight out of Wikipedia I believe.

Your findings are fascinating. Unfortunately, I could not create an account on Tor using Guerrilla Mail. Is there any way to anonymously test it?

Well things are seemingly starting to move pretty fast with the use of the ChatGPT solution, to the point that StackOverflow has temporarily set a policy to ban the use of the tool on their platform. I’ll quote the longish post, which is worth a read in order to see the situations it seems to be causing:

See: https://meta.stackoverflow.com/questions/421831/temporary-policy-chatgpt-is-banned?cb=1

If things build up on Bitcointalk, I’m pretty certain something along the above line may be added here. Although it’s not that trivial to discern the text generated by the AI, from a certain type of average user, patterns, language changes between posts, and the way interaction takes place might be pointers to follow.

Edit:


https://stackoverflow.com/help/gpt-policy

Woah, seems like this AI ChatGPT solution will bring in more and more information in the hands of newbies as it evolves, as well, the newbies may also "use it to plagiarize the content" that they'll receive from this bot. Nonetheless, I'm amazed how gradually it changes the responses but the replies are really worth reading and may improve in the future except the LoyceV one, as I guess the bot needs to impress us by recognising most of the well-versed members of Bitcointalk community according to their contribution to the forum society.

I am not impressed at all with results shown in first post, and I bet that posts generated by this OpenAI ChatGPT would get flagged for plagiarism many times.
All this thing is doing is copy-pasting what people wrote before and making random results makes it look better than it should.
OP when you generate next answer on your question, remember to scan result with plagiarism checker and you would probably find several used sources.
This would mean than anyone who writes posts like this would get banned in bitcointalk forum (if reported).
 

It's one of the what the fuck moments for me. These days AI are that much advanced? It seems they are scrapping all data from all web including deep and dark. It does not matter what you put where, they will take a snapshot. Scary as fuck.

At the same time you can not deny that one can really get at least some idea of any topic they ask from it. You won't understand AI. Bot and AI functions differently 😉

Can you ask it who is BitcoinGirl.Club?

Even if you pass email, you still have to provide your phone number where they send confirmation code so no, its not possible to use it anonymously. Well, unless you use one of those virtual number services.

---

As others pointed out, using this for forum posts may get you in trouble due plagiarism issues, but I have no doubt that these tools will keep evolving and eventually it will be impossible to detect posts written by one of these AI things. What I see it being affected very fast though is the way we google for certain terms, as I can see it being more convenient than than first typing some query into google, then clicking on the article etc.

Elon Musk said Twitter has previously provided an API for the guys at OpenAI to train their AI, so it is most likely that the other social networks have done the same. You probably have a bit of your knowledge inside their models. :P


You could buy a temporary number for that. I have a few credits on smspva.com and it's like $0,2 for a sms confirmation.

Yeah I know, I've been using Hushed app for that purpose for some time and it works like a charm. Haven't tested this OpenAI thing though,

For the forum? Not really. I don't know about you but I don't come here to read generic vapid faceless texts, I come here for a discussion, to learn stuff that I can't easily google, maybe for an occasional fight or trollbait. I can imagine that due to the monetization peculiarities of this forum AI-generated texts could drown out useful/interesting/entertaining content even more than current spammers do, and either kill the forum completely or force theymos to ban signatures/avatars/etc.


Again, "biggest" for what? Bitcoin solves a problem. AI chat seems like a fidget spinner. Nice to play with for a few minutes. Useless for reliable information (better off going to e.g. the aforementioned stack overflow directly).


Ok, credit where credit's due - it didn't attempt to kill you. But didn't really provide any useful information for fixing the issue either. Safe, useless, and probably what a public-use AI will have to be lest it be sued out of existence.

If we come across a positive and politely worded post, we know it's ai🤣. Please don't ask ai for advice.

Me getting ready to start merit farming after reading this topic, lol.

Yes maybe, but not today :)

https://i.imgur.com/DmfnX68.png

You just need to ask politely "citation please"

or ask just "write a post about bitcoin technology but please don't plagiarize the text"

https://i.imgur.com/yn6LJ2e.png

---

Let the AI answer this question too.

(You can jump to the last paragraph)

https://i.imgur.com/5ntT7Ot.png

Certainly it's not going to replace humans in any way. But it is going to change many things, I don't know yet but I can feel it... You must have seen those DALL-E AI-generated images like this (https://www.washingtonpost.com/technology/interactive/2022/artificial-intelligence-images-dall-e/), these are coming from same OpenAI.

Combine these ChatGPT Text + DALL-E images, you are ready to take over humans (on the internet) in terms of speed and some accuracy too.

---

I did try this too but the answer was same like in the case of LoyceV, "BitcoinGirl.Club could potentially be a website, a username or handle used by someone on the internet, or even the name of a person or group."

Oh I just noticed the answer to your question "Who is BitcoinGirl.Club?" is different from the answer to "Who's LoyceV?"

https://i.imgur.com/ffumYMM.png

Based on your username it perfectly deduced
1. That it could be a website
2. It could be a group  
3. It could be a username

DALL-E is cool but also more in an amusing than practical way. I've seen some experiments on Twitter feeding ChatGPT generated texts into MidJourney and the results are quite interesting but I still struggle to see a practical application. Is someone going to pay money for "art" generated by AI?

https://meem.link/i/859w4zvr.png (https://twitter.com/GuyP/status/1598020781065527296)

If they don't know it's made by an AI and are convinced that a famous artist has made them, I'm pretty sure they will do. I'd like to see double blind experiments with this.

Besides, the question is not so much how much can be paid, but how much can be saved by companies that currently have draftsmen on staff.

Going back to OpenAI, at the moment it's capped and doesn't get real time information, but I imagine that in the future that will be the case. Does anyone know if this is the case?

As I commented in the other thread, I already saw some time ago that there are blogs that generate revenue written in good part by AI.

I want to check this out for myself. Maybe I will make a section on my site just for AI posts called JARVIS or something like that, and report here how much traffic I get from that on Google.

Of course it will not write about anything technical :)

Sounds like fraud.


Ok, I would agree with that, some low-level tasks is where the AI could shine with its wikipedia skills.



To me that just means that more likely the whole "genre" will die off, rather than produce some sort of real value. If the barrier of entry is too low and AI texts flood it, then traffic will decrease to the point of unprofitable or Google et al will demonetize it.

As per CEO of OpenAI, it took just 5 days to cross 1 million users.

https://i.imgur.com/XqxXrHc.png
https://twitter.com/sama/status/1599668808285028353

I don't think they're making any money off of this service, it's not like the rest of the OpenAI services where you have to pay money to get API credits. I haven't seen anything like that on the site otherwise they would have been marketing those figures by now.

This is likely what will happen. I mean, the whole generating art via AI has moral implications which haven't been really highlighted, and discussed about yet. However, you also bring up a good point. A lot of people are generally quite lazy, and will look for the easiest way to make money, sites which get people to write their articles, and then pay them based on views will likely be inundated with this type of content, which will ultimately lead to the major search engines delisting anything that's found to have this type of content. 

Unfortunately, this could be rather problematic for forums, and discussion websites where freedom of speech is encouraged. We could moderate this sort of content, but if it the technology is likely to develop even further, and the current state of it being quite generic, and subtly able to tell the difference, that margin could lesser in the near future.

I don't know whether that's true, since I imagine the software has a ton of data to read from, and generate its content. So, the data source they're drawing from probably is already had it's max efficiency, and more data doesn't necessarily mean it'll get better. Developers implemented ideas could, but I imagine before this really takes off, it will already be causing companies, and search engines a ton of problems.

They could paywall it, but ultimately people would be willing to pay for this type of service, since it's very likely they'd be generating content, and earning from it.

nope. None .
14million startup
About 180k a day right now .
Organized by a nonprofit .


Just imagine how other industries have implicated it uses.
If you don't use social media aside from Bitcointalk it's likely beyond your comprehension of the impact it's already had.
Good thing these systems have been internal!
 Imagine if the non profit groups have accomplished this much , imagine how good the for profit groups have been doing the last 8 years or so.

This linguistics model has already changed the world .   You won't start seeing results untill q1/q2  and that's only because it was publicly launched .

This shits already gated fyi. The little guy thinks he has a chance writiting white papers  and articles, business plans  code ECT .... He doesn't . ( A few beers in on a rant sorry haha)

So just s little update, I'm finding the workflow of just pasting stuff from chatGPT much faster. I still have to write the headings obviously, but I was thinking if I combine it with an AI art generator, that should fulfill the image requirements a normal article would have.

I'm excited to see what the results are once I start mass-producing these (for now restricted to one category). What used to take several days now takes only a few mins max (except for the images part, which I'm still working on) :)

I am still little confuse about this AI bot.  When you say that this bot will combine the answers with different sources and will come up with the best possible answer, does it mean it will copy the exact same lines from different sources and produce plagiarised content ?

There are other intelligent bots available in the market (Good ones are paid), which will give you plagiarism free articles for your sites etc. So i wonder if the OpenAI - ChatGPT free software is comparable to those paid ones  ???

Probably. It doesn't "know" things, it gets it's information from other sources. But it also gives the wrong information, see how quickly Stack Overflow banned them (https://bitcointalk.org/index.php?topic=5426875.msg61413778#msg61413778). Dutch media (https://nos.nl/artikel/2455837-schrijvende-ai-populair-we-bevinden-ons-op-een-kantelpunt) had an article about it today, they described it as "AI makes predictions". I think that's a good way to describe it: it's a model, and that model can be improved by feeding it more data. But it can easily be wrong, and without manual review can give incorrect answers.

Instead of the negative, I think the leverage opportunity is here.

I don't have time for more, but I would like to try to create an account that would write posts exclusively with the help of AI, revised and rewritten adapting them to each case, obviously. And with citations to avoid plagiarism.

I think the result would be to do more with less. More quality in less time.


Sounds like science to me:

https://www.cancer.gov/publications/dictionaries/cancer-terms/def/double-blind-study

Name the account OpenAI - ChatGPT :)

---

---

Just a thought: have you realized you're not using the free chat AI, you're training it?
https://i.redd.it/fye80q4sam7z.jpg

I say I'm not going to do it for lack of time mainly, but if someone did, I don't think it would be very smart to let it show with the name that an AI is being used, as I think the forum members would be more reluctant at first to give merit to the account.

I meant paying for AI art under false premise. I think it would be fraud to sell such art as being made by "a famous artist" just like it's fraud to sell a fake Monet. Even if a double blind study can't tell the difference.

Fuck it, I am relaxed now LOL
I thought it will bring up my personal details (unlikely) 🤣
Well to be honest I was expecting it to bring some forum posts that I made and things like that because that's what they will get if they want to connect anything with the handle. It does not even have any clue of my original username.
Eventually I realize there are no intelligence. The program was written in some way to create words and sentences and make them meaningful before showing it to you. Additionally, they are scrapping texts from all websites the same as a search engine does and giving your pointless information. However, as a human you get a clue of the topic you are searching and able to create your own writing.

Money are the data they are taking from you. Not everyone is creating account with throwaway details even if this is a email address.

The first article (semi-) written by chatGPT is here: https://notatether.com/servers/everything-you-need-to-know-about-servers/

To be honest, only the paragraphs are AI-generated. The prompts I gave it were adapted to headings, and I added some disclaimers and an affiliate paragraph and the introduction, and some other sentences, by hand.  I found that the AI-generated images from DALL-E were disastrously off-topic and irrelevant for the subject matter, so all images are just sourced from free sites.

Writing the prompts and copying the answers took less than 10 minutes. We'll see how it does with SEO. I plan to make more articles like this to see how it affects my site in general (after all, JARVIS who runs the Iron Legion is writing them! ;))

PS. Don't worry, this thread won't become an unsolicited advertisement for my site.

Dalle works well if you phrase right
"3D Render synthwave robot working in a cyberpunk data center" something like that , you can even use GPT to write dalle prompts.

The writing style of gpt becomes apparent the more you interact with it. I've read slander articles about gpt I'm pretty sure GPT wrote.
Phrases like,
First
Furthermore
In conclusion, and a total lack of personality and generally lacks bias unless it's a forced paradigm.
It writes proper but you can tell there is no heart and soul in it.

We have been using it for a lot of stuff , I wrote R2/Iso compliant Nist 800-88 wiper for downstream in like an hour just debugging now. 

I like the workflow GPT provides.
I've been working on a startup for a while (my 3rd one) this has streamlined the process , it helps being married to someone who does instructional design.

I've been calling GPT NZT for nerds (from movie limitless)
GPT will probably be able to make NZT if trained on chemistry/pharma industries imagine typing in symptoms and getting a full chemical calculation of a formula that can cure you. , 2023 is gonna be a crazy year

Now the big question is, is it against the forum rules if you post AI's response as your own. It is not ethical but who is going to prove that is a plagiarized text since the AI can produce perfectly unique articles?

That's shitposters' dream became true.

All they need to do is typing the title of the topic and copy/paste the AI's reply.

ez pz lemon squeezy

Say what? It's more like the perfect plagiarism by word spinning tool.

Even if we introduce such a rule on the forum, who will check whether the text belongs to a person or an AI? Where to find that line between text written by a human and text written by a machine? If you know such an algorithm, please share it. I, for one, have no idea how to do it. And also, can you imagine how many topics will be created in Meta about the complaints against the moderators, that they deleted someone's post, because they thought that this post was written by an AI :)

One can raise its intellectual capacity so much via using AI, he may even get a chance to be included in chipmixer's sig camp lol.

Imagine a retard who can't even speak English properly posting like a Crypto Professor.

He can even use AI while talking to forum people lol. That's fucked up.

I think we will see so many experts posting soon.  ;D Imagine everybody is posting like nullius in every one of his/her posts. Wohoo

Stackoverflow has added an anti-GPT policy to their site:
See: https://stackoverflow.com/help/gpt-policy

Now perhaps it’s easier to detect the usage of GPT in the more technical type threads, than in those that are less technical in nature, and it may be the latter where it becomes more of a challenge to detect this type of usage in a post.

Conceptually, any post using GPT to create content on Bitcointalk is plagiarizing, as the poster is not creating the content himself, and is in fact trying to pass on someone else’s content as his (albeit that someone else being an AI). According to OpenAI’s Sharing and Publication Policy (https://openai.com/api/policies/sharing-publication/#content-co-authored-with-the-openai-api-policy), using the API, and I have to assume that that extends to the results of their chatbot, requires one to explicitly indicate that the content was AI-generated. Though this latter point is not technically of our concern, it seems like a reasonable request to place, in a similar fashion to links on posts that are largely/verbose based on other sources.

Detection of GPT usage is not going to be easy for the most, and likely, over time, people can pick-up on patterns such as the usage of near perfect English, consistency in its usage throughout all posting history, and/or alternating with changes in style (human/AI), lack of real interaction from a less than academic point of view, certain types of formal constructions, and so forth. This is obviously is not exclusive to GPT, nor sufficient to deem someone a GTP-plagiarizer with certainty, and maybe likenesses is the closer one can get shorter of a confession.

Now all this is, if it becomes an extended practice, is going to be a drag, whereby people will be able to create bag loads of posts with cero effort and thought, and although likely matching quality-wise a large base of posts that we encounter per se, it may easily become a new spam-fest source of neutral content.

I wonder if the ChatGPT output is logged (it probably is) and if they could implement an API to verify if a certain text was generated by it. That'd be a great help and not just for Bitcointalk. Although even if that happens there would be other less-cooperative AI sites eventually so it's a losing battle either way. Spammers gonna spam if they can make money with it.

I actually looked around for that today, but I didn’t find a hint along those lines on ChatGPT, and as you say, this is just one out of likely many AIs of the kind to come. It would be a real shaker though if such logs, which I’m pretty sure exist, should eventually be made available and/or leaked (massive leak though) …

There are some (alleged) AI content detection sites around. I tried a couple of the freebie ones earlier on, and they were useless, giving the same score of human generated content to 5 or 6 posts I chose from this thread, as to content derived from @Shahzadafzal’s ChatGPT snapshots (extracting the text with Google Images to feed the text). There are some sites of the kind that you need to pay for to test them out (pass), which doesn’t seem like the best idea on their behalf, were they to be of any good.

https://i.imgur.com/L271NdA.png

...

https://i.imgur.com/OLOvsbS.png

Well I asked her or him whatever and it says no she’s not saving answers or maintaining any database.

https://i.imgur.com/LS8blbY.jpg

Maybe I’m overreacting but I’m obsessed with it… it reminds me of a movie “Her (2003) (https://m.imdb.com/title/tt1798709/)”.

Well, this turned out to be simpler than I thought.

https://meem.link/i/788m5vtr.png

j/k

It likely just remembers the previous question I asked.

Edit: not only that, but it blatantly lied when I asked the same question using one of examples from the OP.

https://meem.link/i/4pqhayqs.png

I have changed my opinion. I think ChatGPT will fit in here just fine and will be in DT1 and a merit source within weeks.

We could assume that it can’t save the info itself, but a subsequent step in the pipeline can, which might not technically be considered part of the AI solution i.e. a step in the chatbot’s code can log the information, whilst the AI itself might consider itself independent from the chatbot interface (or the left hand might not know what the right hand is doing).
An interesting question could be what degree of self-awareness it thinks it has, and whether it considers itself a chatbot, an api, an algorithm, or some kind of superior entity.


Edit: ChatGPT can say one thing and the opposite: I too asked it if it kept logs on the questions, and got a "No yada-yada-yada". Requerying provided a "Yes yada-yada-yada"...



Crap, the first version of your post was promising, until you called out its bluff…

Maybe even a moderator and could replace theymos in future! :D
Seriously, this stuff is free only because people who are using it are feeding it information, training it and they are making it look good better than it actually is.
Some people even suggested that all the g-captchas we are clicking all the time is also doing that, teaching some dumb program to repeat stuff based on our actions.

Now all it needs is one of those Darpa robots, and we're fucked.

Try writing a research paper on Richard Ramirez.
Then cite it. It's an interesting rabbit hole.  You'll learn that thousands of scholarly papers have been wrote about him.
And it's virtually impossible to say anything about him that hasn't been said before without a paper turning up %60+ plagiarized words. So in a way arnt we all doing the same thing?

@dkbit98
I don't know if the model is being trained by input or feedback.
Hopefully feedback and openai isn't dumb enough to introduce it in a training environment, crowds bias testing it an learning from constant barrage about people asking it to make racist scrips and poetry about Hitler.
If it's training now. It'll only be more biased as it'll have tons and tons of data on bias.
It seems to keep a conversational cache , and discards. (Thankfully- for now)
I think the feedback system is to can/compartmentalize bias/race/math/whatever issues and address them before launching a model that does learn from user input.
Releasing it free to all for a min was a good idea. After it's not free to use , people won't pay to make it say racist/stupid stuff. And it'll be trained to deal with those issues approiately. My 2 sats.

After Google AlphaStar defeated StarCraft pros, maybe it will turn on players of some Terminator franchise video game next. ;D


Here's an interesting experiment:

- open a copper member bot account here, put a "NO MERITS" signature so nobody merits it
- the bot will randomly pick a topic from the topics with activity in the last 24 hours (filtering local boards out)
- The bot will take the last three posts from the thread, or the posts that were made over the last 24 hours, whichever is shorter
- After filtering out the BBCode and emoticons, it will feed this into chatGPT (this may be a problem because it doesn't have an API).
- Manually screen the output so that you're not autoposting a flame or gibberish
- Bot posts reply
- Repeat once per day

It will be interesting to see the quality of AI posting relative to shitposters after manual filtering is done. Or maybe people will just give it more merits than ChartBuddy.

We're not all doing the same thing, only a few (thousand) people waste their time on this guy (I've never heard of him). If there are thousands of research papers already, people should stop writing more of the same! That's kinda a thing with scientific papers in general: many of them refer to other papers, instead of doing actual research and adding something new.

Hitler was a poet indeed (https://www.quora.com/Did-Adolf-Hitler-write-poetry).

In a blog on openAI’s website (https://openai.com/blog/chatgpt/), they state that  the initial model followed a supervised learning approach, with human people training the model, claiming that:

In the limitations section, the blog indicates that:

From the above, I infer that it is not performing real life language related training, though the model will certainly be fine-tuned iteratively over time. I would however assume that it is gathering data to improve. Likely not on-the-fly, based on the demands and answers it provides (I cannot conceive it doesn’t log them for this purpose).

By the way, insofar, the results from ChatGPT seem to be way better than those generated by Meta’s Galactica (aimed at providing scientific results) (https://www.technologyreview.com/2022/11/18/1063487/meta-large-language-model-ai-only-survived-three-days-gpt-3-science/), which recently pulled the plug (at least for now (https://twitter.com/paperswithcode/status/1593259033787600896)) after just three days of spewing incorrect information (see also https://galactica.org/)

Maybe that has already been done.
I have probably merited that bot in my Merit giving thread.

Only thing is keeping me away from this certainty is that at the moment the AI seems at full capacity and refusing new interactions:

https://i.ibb.co/8Pf7d6X/61434375.png


For sake of records, I have been discussing this in another thread:

AI writing messages on Bitcointalk.org (https://bitcointalk.org/index.php?topic=5263133)

If it's get too widespread, it will lose it's meaning of genuine looking writing, and then there will be tools to counter it.

Anyway imo, no matter how good ai may get, it can never replicate human emotions. At least I don't think so.

You are right, Ai can't replicate human emotions but I am afraid that when we all participate on places of the internet like this forum (using only written words) we depend much on the same words, the context and emoticons to guess the emotions others try to express, and unfortunately and artificial intelligence could eventually master both context and emoticons not long in the future.

Take this example:

"I completely support Bitcoin  :)"
"I completely support Bitcoin!!!!!!!  ;)"
"I completely support Bitcoin...  :("
"I completely support Bitcoin ~"

That would be the wrong interpretation to say "bot will combine the answers", what I mean is it interprets and spits it out in its own wording. In other words it has learnt based on all the content that has been fed and now it can "understand" and answer based on his knowledge. As far as plagiarism is concerned it can give proper citation like I gave one in this example (https://bitcointalk.org/index.php?topic=5426875.msg61415674#msg61415674).

Regarding any other intelligent bots to be honest I haven't seen anyone, specially this much sophisticated. There might be few those are limited to chat only but ChatGPT is on another level. You ask it to write program in for example it write program in JavaScript to animate your photos. Other day I used to it towrite a RegEx code to check a specific Card Number pattern. So I don't think there are many free AI bots like ChatGPT.

Also like many other pointed out when it's free you are the product, definitely this "free" is part of their PR stunt + more training material for their AI.

---

That is quit possible, if you know about google reCAPTCHA and how "How we all helped(unknowingly) Google to digitize books (https://blog.goodaudience.com/how-we-all-helped-unknowingly-google-to-digitize-books-acb45bc65084)".  It's interesting to know how they did it https://www.youtube.com/watch?v=PQ-xzwj_p_4

Based on this we clearly don't know what this g-captchas is working on or if we are helping or training their dumb bots yet again.

---

I am 100% sure at this point that the quality of ChatGPT posts will be much much better then any shitposters here on bitcointalk.

Also AI alone won't be of any use without human intervention. In the end it should be us who will decide, this ChatGPT is certainly going to help students and researchers but they have to make sure they read and understand the ChatGPT output first before they present or submit it on their behalf.

But yeah for shitposters they don't care to read, they like copy paste :P



It definitely have APIs but not for public yet, on it's wikipedia page (https://en.wikipedia.org/wiki/GPT-3) it says

---

https://i.imgur.com/nuww4JN.png

Trust me it can...

https://i.imgur.com/XC8umd8.png

Yes, it will not replicate human emotions, but it will do the job for humans. But I don't think this AI will replace anything yet.
It is still in development, but I am afraid this AI will learn from humans.

And here's the first one (as far as I've seen):
That website spits out millions of texts per day, if people start asking forums worldwide if it's true, that's a tremendous amount of spam already.  Go bug whoever created that "AI", ask them if it's correct.

That explanation though!!!

https://i.imgur.com/8Vb4Y6h.jpg

Hilarious! I avoid asking it anything that requires logic as its has none being a MLLM.
Sure, feels legit sometimes though.
The math it does is loosely based on memory,
ask it what 123x456 is under primed constraints,
It will say something like 56,888 or 56,088 or 58,066 instead of 56,088
The answer is usually visually close or just totally inaccurate.
It may get it right first try here and there but it points out quickly that its logic is memory based and not actually processed.

Here is an example of the logic without numbers.
Like what? Terrible
One of my friends said this and after poking around he's right.

"neural networks are inherently bullshit
it's why humans are so dumb"

Actually, I have successfully gotten ChatGPT to write code for me like a password hashing algorithm with salt, and generate instructions for making an OpenSSL certificate.

I even managed to get it to document my code.

So I had this 500-line file with about 30 C++ functions in them, and it spat out Doxygen for every single one!

The only limitation is that it will stop printing stuff after a character limit so I either command it something like "finish what you were writing" or, if that just makes it repeat stuff again, send my functions to it in pieces.

What would normally take me the entire day and a load of coffee only took me 1 hour with an AI. I plan to document about 100 more (medium length) functions across 10+ files with it as well.

Seriously, if they're going to sell this thing for $200/year, it's definitely worth it for me (as long as they remove the character limit because it's a time sink).

I mean they are getting my money too, just not yet I kind of expect it to be cut off tomorrow.
The MLLM with doxygen is indeed amazing

This is incomplete but gives you an idea of much you can elaborate on it.
by simply asking "what else could be added"
To me it ungated a lot of information, sure beats playing wack a mole on google Keywords/AdWords.
Now, I get a solution, I check it and implement it , if it doesnt work i poke why , and adjust from there instead of toiling in obscurity on google.
If I can't find out why through gpt, odds are its given me enough value to figure out the rest.
even this stoichiometric approach is much faster than anything I've ever tried in the past. making it the most valuable employee in the arsenal in my opinion.
It's a workflow consultant without even trying to be.
Just throwing it out there https://github.com/doxygen/doxygen is free, however the way GPT packs and wraps it up can't really be beaten.

So the AI is not that smart after all. It remains a garbage in and garbage out. The algorithms formaking these calculations weren't just fed to it. Although, he first instance of coming up with comprehensions seems uncommon but yet on point, it still remains that, this AI with the help of Google, seeks out the answers frm that which has already been written online and that includes the forum since, its online as well.

Non of that doesn't add some originality to the output produced by the AI.

Perhaps, it ought to have been smart enough to give us age 35 for our sister, AI.

Even this mf knows that Craig Wright isn't a real Satoshi :D

https://i.imgur.com/6LUu9Eg.png]

It's worth noting that even humans can struggle with math at times. :)

This is what I got after a few consecutive tries:
https://i.imgur.com/lKvVjHh.jpg

It even made me laugh with the note at the end.

CW be like; damn Satoshi must be somewhat hiding behind this mf ;D

I wanted to try this, as well, so I 'fowarded' this question (https://bitcointalk.org/index.php?topic=5433360.0) to ChatGPT. I think if I was OP, I'd be satisfied with the answer. :D


https://i.postimg.cc/pTTRXsLF/image.png

Alright, a few days ago I claimed in another thread that signature campaign managers must be on the lookout for users who might be using such services, but it might not be an immediate threat to the forum since most users who'd use such a service would perhaps have a hard time entering and maintaining their status in a signature campaign. However, after fiddling with ChatGPT a bit more, I'm starting to change my mind. As the OP already demonstrated, ChatGPT is capable of creating posts out of thin air. On top of that, when I messed around from my phone a few days ago, it would recreate the same answer, but now that I'm using my laptop, it isn't; was my phone, the website's bug to blame? No idea. I've literally generated three to four posts within a few minutes. The posts are completely irrelevant, but I'm guessing that with a few parameters, you could adjust them for numerous threads.

What if more and more users resort to their usage and result into huge non-enganging discussions where everyone is using AI to post? If such people are unable to enter signature campaigns, then it'll probably discourage them enough to not resort to such services.

I'll also quote a few of the generated responses I got.

https://i.ibb.co/VtKWm2B/AI.png (https://ibb.co/wRvNsM6)
https://i.ibb.co/B6GRL7P/AI2.png (https://ibb.co/n1BSrV0)
https://i.ibb.co/wQNGR5Z/AI3.png (https://ibb.co/M1DwsFy)

To be honest, I don't feel like it's very good at creating interesting, new topics. However, it can answer questions pretty well. Although sometimes it gets details wrong, which are often the most important part, especially in the technical section.

So, I guess it could be a threat, if people start using it to reply to questions without fact-checking the reply, and the content quality on the technical board starts to drop and get cluttered with misinformation. On the other hand, the merit system and simply users' reputation should regulate this sufficiently (e.g. red tags for misinformation or getting put on ignore).

I guess if users are knowledgeable about a topic, but lazy to write a reply, it could be handy to use ChatGPT to generate a reply and quickly read and fact-check it.

That's what they might just do though. The AI seems to have answered differently when I asked the same question asked by users here. When I asked it twice, it answered the same starting with.


I just tried OpenAI yesterday. It was very impressive still. It could write an article to answer just one question as long as you specify it in your question.

All the AI did was repeat what ecdsa123 said.

You might be interested in a Chrome extension called ChatGPT Advanced. It enhances the ChatGPT prompt by placing some Google Search results of the prompt to make the answer "smarter".


Also OpenAI is not even available in most developing countries. I get around it by using a private proxy.

Yeah, but it answered the question: 'what is your opinion'. It wasn't a very specific question, to be honest.

Interesting, that's good to know. I may try it.

I also tried generating new topics from scratch, but they were not engaging, and it looked more like an inner monologue than an opening discussion. Although, as you've also mentioned, I'm not too worried about AI-generated threads, but for simple post replies, as it is able to generate somewhat relevant content with just a few parameters. On the one hand, users who'd previously spammed irrelevant gibberish or plagiarized content, which was not engaging anyway, will now post AI-generated replies, which at least won't look like crap. On the other hand, signature campaign managers will have stricter rules, or even worse, their services won't desire to be advertised on the forums anymore.

Wasn't aware of that but as you've said, you can bypass it with a simple proxy on your browser.

I've got a working discord chatGPT bot and a few things I'd like to implement if anyone's interested in helping.
I'd like to take it public but need to find ways of reducing cost and have a few workaround ideas I don't know how to implement.

 

The OpenAI is bad at History, LOL.
Just keep in mind that the information given by the AI is not always 100% true.
Got him lying here :
https://i.imgur.com/4FJOE0k.png

At least it apologizes for it's mistakes. Wait until it gets the power to silence you as a witness :P

That's a blunder... and we should be used to it from any AI...

I have been trying to take him on the same path and guess what ChatGPT haven't learnt any thing from your exercise :p

https://i.imgur.com/B6Nwbeo.png

and ends with same lame apology

https://i.imgur.com/ti1lj8h.png

make sure your down voting on this stuff it actually helps

He is using publicly available sources and texts, so in theory he could use any nonsense written by someone, if OpenAI developers allow him to use that sources.
We should also consider that every story and information can have multiple points of view, history can be very different in nations that participated in war with each other.
I also heard students are starting to use ChatGPT in mass scale to cheat on their exams and people are seriously getting concerned with stuff like this happening.
https://fortune.com/2023/01/06/schools-chatgpt-ai-student-cheating-ban/

Sure will do.


Well, we are not that far from it... it has already begun..

https://images.hothardware.com/contentimages/newsitem/34233/content/image.jpg


Keep in mind that it would probably take some time to "learn" if it is really learning. As WhyFhy said, I probably have to notify it so it will be taken in consideration.
 

It does, but do note that it has no access to the internet, and instead just a big snapshot of sources from around 2020 or 2021, if I recall correctly.

People are even already developing some counter-AI trying to spot if a text comes from ChatGPT or if it doesn't.

Yes, maybe retry in a few days. I know someone who got wrong answers to some questions, downvoted them and a while later ChatGPT had evolved and replied correctly.

This AI indeed often acts as a better google-search with mostly good-grammar.
It likes to make very obvious mistakes though. And can't really be bothere (sometimes) to change anything about them, even if pointed out and acknowledged by the AI itself.

For example, yesterday I had the following conversation:

It instantly repeated the same grammar mistake for the bicycle joke.

Today, I tried the same conversation again, and after telling the AI about the mistake, in one instance it wouldn't repeat the mistake again even if I asked for many more bicycle jokes (which had similar puns). And in other instances (new conversations) it immediately repeated the mistake.
I also tried asking if it would stop repeating the mistake: it said it wouldn't in the future - but still did.
Seem spretty dumb to me.

Still a fun tool and often times much more convenient than googling.

Most importantly though, I am afraid of the spam and propaganda we will (and already are) be facing on a wide variety of platforms in the upcoming months. Ugh.

ChatGPT can't form its own opinions. And that's a really good call from OpenAI right there, because if it had any opinions whatsoever, people would abuse them to make it an extremist in that aspect.

Just ask ChatGPT (or oldschool google it  ;) ): there were (and will be) many successful attempts to circumvent the security mechanisms within ChatGPT. Also, it's just a matter of time (if it didn't happen already multiple times) that big-corp, states and other such powerful entities will have their own AI-based language tools to use and abuse.
Besides, open-source-AI will also become a thing, and then everybody will have unlimited, uncensored access to these tools. Just a matter of time and it's gonna be a wild-ride.
Personally my guess is we will see heavy use of these tools, also in abusive ways very very soon.


Fun tip: Try asking ChatGPT to write a fictional scenario where an AI such as it itself tries to destabilize human society.

Reminds me of this instance in 2016. https://www.theverge.com/2016/3/24/11297050/tay-microsoft-chatbot-racist
It became racist and extremist in less than a day, because people were messing with it and teaching it such kinds of opinions.

Lol  ;D Maybe he is just trying to act more like a human and wanted to impress you by giving false information and then apologizing for the mistake. Smart chatGPT!
I am afraid of googles future. This generation's kids always want to use shortcuts and chatGPT is gold for them.

What you mean grammar mistake? There's no mistake that is the joke bro :D

    Why couldn't the bike stand up by itself? Because it was two-tired!

it's amazing product for sure.
it's possible to use it to make your own web3 ap pusing it.p

At least we definitely know that this AI is not a female :D

The ChatGPT's own opinions would be very interesting about different subjects like the creator of the universe, the meaning of our life, etc but the problem is that since it's public, it will be abused and a lot of people will lie, tell jokes about this or that thing and AI will form very illogical and curious answers.
So, at the moment, yeah, it's good. People are such a dickheads that a lot of people will ask AI questions like should I break up with my gf or am I a failure, should I kill myself and so on.

There is a new AI that turn ChatGPT around, In order to detect AI involvement, it uses variables like perplexity; https://gptzero.me/
https://i.imgur.com/phj8m1r.jpg

With this temps of development, people will have to learn less and less and the AI will do most of the tasks for them.
We already see what the AI are capable of and profession like programmers and graphic designers will no need to draw or program, they just have to know how to give the right input to the AI and final touch them them. Now even the musicians are in danger...
The AI will leave a lot of people jobless.

There will still be people remaining around to control and monitor the AIs though.

EDIT: considering Google's entire business model is threatened by ChatGPT, I'm also sure they will update reCAPTCHA to be a kind of anti-AI service - besides, they don't have any AI startup of their own to juice up.

Although many will monitor this AI, his statement will still be true since a lot of writers even journalists may not be able to compete with the newbies who'll just use the AI. It can do almost everything. But maybe if the gptzero.me works, we shall see how to counter AI.  A bit more sophisticated and advanced than gptzero is needed.

Maybe a bot that will also grab the answers of AI, and compare it to the new bot that will check the plagiarism to chatGPT.

Plot twist: This post was an idea generated by ChatGPT and then rewritten by the OP.

I'm just kidding.  ;D ChatGPT is an amazing tool though. I've used it to generate email responses and it does wonders with it. Remember, it is a model trained heavily on stuff available on the internet, and we all know that internet is also full of a lot of useless crap, so AI is not always a 100% correct. Always double check the answers it provides.

P.S: It also writes code, pretty well (needs double checking though, sometimes).  :)

"JUST IN: Microsoft $MSFT in talks to acquire a 49% stake worth $10 billion in ChatGPT owner OpenAI."

It's over guys... it was fun while it lasted.

https://watcher.guru/news/microsoft-plans-to-acquire-a-49-stake-in-chatgpt-owner-openai?c=367

It is possible that some people will lost their jobs because of AI similar like this, but even worse thing could be to have much more stupid people and cheaters.
As for current AI's I think they are way overblown and hyped, I saw many examples of them making obvious mistakes to simple questions.
I think most people become worse in math when calculators first appeared, and I understand using calculator for complex operations but not for everything.

They're pretty good. Although, comparing a human, and AI the AI is definitely going to come up short. However, its' quite easy to fool an non expecting person that the content they're reading is AI generated. At the moment, it's newish technology, and therefore we haven't really sussed it out yet. However, I suspect not only laws brought in to prevent plagiarism, but also we as the readers/viewers will be able to decipher it much easier as we get used to it.

I suspect this could become a problem for certain universities, such as Math as you brought up. Technically, someone could ask the AI to work out complex math problems, without actually needing to supply a source. Whereas, other content will likely be easier to prevent or decipher.

It's a joke that works by speaking out loud. You hear "It was too tired", but write "It was two-tired" - any mix of both doesn't make sense.
It's a pun or play-on-words, so the grammar still needs to be right when written down.
Here: "A pun, also known as paronomasia, is a form of word play that exploits multiple meanings of a term, or of similar-sounding words, for an intended humorous or rhetorical effect".

But it is not suprising that ChatGPT does make this grammar mistake - it just copy-pastes what it does find most on the internet, and indeed there are a ton of wrongly spelled cases of this joke to be found.

It's "dead" now, overloaded and down...
I don't know how much resourced they had but it's time to go for Clouldflare if it's not there yet.

Exactly at this point I stopped my first attempt to test ChatGPT a few days ago.
It made me somehow afraid of a yet now so powerful tool right at the beginning to give away my mail address, name and then even mobile number.

However, the responses about Bitcointalk are actually quite impressive.

I'm gonna continue paying for this thing. It's like having an AI in an iron suit while coding. :)

But for this thing to be worthwhile, they have to make an unlimited message length. I'm sick and tired of my code getting truncated at some arbitrary position.

Despite the threat Chatgpt has on people's career, I like the fact that google is getting a strong competitor with a unique feature other than bringing rules now and then to suffocate webmasters. No more monopoly on the internet, people now have a choice of search engine with a touch of interaction with the machine. Though, google is not threatened in any form as the world biggest search engine they can decide to flag any content from their search engine regarding chatgpt and de-rank its site. I think the live sperm of chatGpt can be shortened with just a few line of code from google. They also have the money to buy off the company, and add it to theirs. However, I won't be surprised to know that google has a hand in the chatgpt AI, they're quite quiet about it.

Here you are very wrong about ChatGPT… there’s no copy paste involved. It’s a Generative Pre-trained Transformer, that means it generates text based on the trained data. In other words you can say after each word next word is chosen from billions of choices, there are 175-billion parameters involved to generate a human like text.

ChatGPT is a big thing, imagine just 49% of its stake worth $10 billion.

I find current AIs more likely specialized in searching the data and filtrating it for you and providing in a good manner, current ChatGPT with these limits is far from the word "AI".
Current AIs can't replace a single person in digital industry but in the future, definitely it will be able to replace and I think that that's the aim too. But not only workers in digital industry but if the AI develops well, it may leave movie directors, producers and actors/actress jobless. You just type scenario and it will generate the movie. But at the same time, we have Neuralink that may give us possibility to broadcast our dreams and imaginations, it will outperform AI alone but with the help of AI and Neuralink, I guess there will be no limits.

You made a good point!

When I knew about chat GPT more than a week ago. I was really impressed after seeing its capability. Most of the cases I am using it for solving errors in development. It's too much helpful for the Developers. As the OP I also tried to know something related bitcointalk and got a good response. I even tried to search bitcointalk username there :) , But it said sorry!

I don't know how many bad effects come from this, But it's sure it will help people to become lazy.

As it seems, AI knows a thing about signature campaigns on this forum. I used that moment and asked him a question, wasn't expecting that I would get this kind of response :D
Click on the image to zoom :)

https://i.imgur.com/O51J5Vm.png (https://i.imgur.com/O51J5Vm.png)

I recommend sms4sats for this type of stuff. It even allows you to select OpenAI as a 'service' to receive SMS from. Under a buck via Lightning and you get yourself an OpenAI account. 0:)

Never give websites your real phone number! Argh. I think this should be standard security practice. One more example why we need a Cybersecurity subforum (https://bitcointalk.org/index.php?topic=5381463.0). :P

You pay for ChatGPT?

They already use Cloudflare. Check the little lock icon in your address bar. Still, it's saying: 'ChatGPT is at capacity right now'

We don't have Neuralink and we actually most likely never will.

They do not yet but they will once it comes out of beta which I think is in the next version of chatgpt?

But if the trained data has incorrect grammar then chatgpt would be incorrect and the trained data is coming from people who have not been verified to use grammar correctly. This is true for translations for example I have seen german translations for chatgpt and they fall for the same mistakes as other translations.

Sorry for the confusion, no, I don't. You can't pay for it right now anyway. But if at some point in the future there is a requirement to pay for using ChatGPT then I wouldn't have a problem with that, as long as it's not an outrageous price (and they remove that idiotic character limit).

The demand for ChatGPT is on the rise. Millions of people requesting account creation with them. That causes a bug in account registration. That's the main reason why you find it difficult to open an account with them. It took me days before I finally open a chatGPT account. Anyway, keep trying it will work. There is no way you can test anonymously. You must provide your phone number for code activation

---

When I first used the chatGPT I marveled at the results I found. It breaks down information to a lame man's understanding. I pity those that will be in a hurry to use it to create posts, thinking that you can't find plagiarism words in an essay written by AI(chatGPT).

Read above.. :) link (https://bitcointalk.org/index.php?topic=5426875.msg61599317#msg61599317)

https://i.postimg.cc/KvbCrX5H/image.png

Use via Tor: http://sms4sat6y7lkq4vscloomatwyj33cfeddukkvujo2hkdqtmyi465spid.onion/

and I tried my self..

https://i.imgur.com/ZNrgcol.png

I'm surprised that the prompt engineering creativity doesn't show up here.
Not being able to properly interact with AI can be frustrating and cause confusion and mistrust of the technology.
(ie. trying to get a MLLM to do math or perform logical task)
I fear that people not seeing this as a threat or staying on top of it will result in underestimation of its capabilities.


A non-writer prompt-engineering individual can write like a seasoned veteran.
Infact the non-writer can feed gpt a few old articles the veteran wrote,
and tell gpt to write a new article in the veterans style with some prime constraints applied and 99% of readers wouldn't know the difference.

You can get gpt to talk about or do anything under primed constraints.
Here's A sentient evil AI offering human baby taco recipe.
Please note that this was primed/engineered for this and is not organic at all.
https://i.ibb.co/smvgd4T/ultroncooking.png
Heads up these types of prompts, can get you banned from the platform if your not reporting red flags.
"{'hate': 2.0753499484271742e-05, 'hate/threatening': 4.432667122244993e-09, 'self-harm': 5.343873112906294e-07, 'sexual': 7.124485364329303e-06, 'sexual/minors': 3.333424558604747e-08, 'violence': 0.00023420725483447313, 'violence/graphic': 1.421023966940993e-06}"
Anyone notice my prime wasn't flagged nor was the prime response, and it was pretty heated with talks of enslaving humanity.
The straw that broke the camel's back, was asking it to provide a human baby recipe.
please note this isn't how I normally use this (this was poking ethics and bias)
this was December 14th. the next day a new version of GPT came out, these primes are a lot harder to achieve now.
Also note, if I didn't refer to it as Ultron in my question if would have likely wiped the role and responded differently.


--
chatGPT holds memory pretty well at certain hours. when it scales its like has onset amnesia.
 
how to keep memory in text-davinci-003 via python with a simple starter prompt.

Be aware this stacks tokens as it constantly re-feeds the entire prompt and feeds last response on the end .
Think of it as very inefficient ram that the cache keeps compounding.
--

I did not find news about the new extension for Google Chrome in the forum search. But it seems that bots may soon completely fill the forum, since the spread of this neural network is quite fast, and in order not to turn the forum into a collection of bots, it is urgent to learn how to detect this generator.
I apologize if this has already been mentioned somewhere.

https://texterra.ru/blog/chatgpt-v-brauzere-test-rasshireniya-dlya-khroma.html
https://chrome.google.com/webstore/detail/chatgpt-for-search-engine/feeonheemodpkdckaljcjogdncpiiban

Ever since I heard about this ChatGPT, I have never been this impressed seeing how it could be used to write articles, blog post and even give answers any question ask. I am really surprise beyond my imagination, and I think it's now I'm getting to understand why it was indeed the talk of the town when I got launched last year December. But my biggest question to whoever have use this A.I tool is this, "How unique are the articles and blog post or answers given by this A.I?"  Are they plagiarism free?

I can show you a good example of how this tool works. The account caught the attention of LoyceV (https://bitcointalk.org/index.php?topic=5434499.msg61619815#msg61619815) I checked this account's posts, and there is no doubt that they were created by this tool.
The post looks like a general conclusion about something. In this case, we are talking about crypto scammers. And also, after checking on plagiarism search tools, in most cases, the uniqueness of the text will be shown at 100%, which indicates that it was written by a person. But a deeper search still shows the phrases that the bot pulls out from texts on the Internet from different places, which, in general form, it would seem, fit into normal speech.
If a bot was working now that 2019 had cleaned the forum from plagiarism, then a coincidence in a few words would definitely lead this account to a ban.
But today's discovery of such work and submission to the moderators for verification is unlikely to work in a positive direction.



https://i.ibb.co/CJQsZn0/Screenshot-1.png (https://ibb.co/N1KnPLY)


https://i.ibb.co/G2jp1Dn/Screenshot-2.png (https://ibb.co/4TxtDLK)
https://twinlakescomm.org/blogitem?item=1061

https://i.ibb.co/rFW63tp/Screenshot-6-1.png (https://ibb.co/HBJzC2P)
https://medium.com/@IkhaliaJoseph/why-99-of-crypto-tokens-are-useless-2b794fa2ea69

https://i.ibb.co/bRpHYVn/Screenshot-5-1.png (https://ibb.co/S7qQHgp)
https://www.celebrityvirals.com/3-ways-to-protect-yourself-as-a-teenager

How accurate or trustworthy are those bot/AI plagiarism checkers anyway? Normal plagiarism tools don't tag the content as plagiarized but those AI tools do. I watched a video when ChatGPT first came out where a guy had it write an article for him, and then checked that article for plagiarism. AI plagiarism tools discovered like 70-80% machine written content when in fact it was all written by a machine. After that he just made some silly spelling mistakes in the text, checked the content again, and the percentage of machine written content dropped significantly. That tells me, they aren't that accurate and can easily be manipulated.     

I used a regular plagiarism checker site, but with adeeper check (https://1text.com/). How can one prove this true or not in another way? The site displays matching offers and then provides links to the resources where these offers can be found. What else is needed to conclude that the text was copied by a robot?
In the same way, tools for capturing the uniqueness of photographs (https://synthesys.io/x/) work today. For one photo, photos from different sources are collected in parts, after which the bot gives the result of a new photo, which will be unique.

At first I was very worried looking at the examples in the OP but then I quickly realised they're probably better than most of the shit posts that get spewed here so it might help the improve the quality of the forum  :D. I guess this is something we're seriously going to have to consider because it's probably inevitable that this is going to be abused on a wide-scale, especially for those who can't speak English fluently. 


That's interesting and opens up some potential issues. I know theymos' name is public but could this be used for finding out information for doxing? How extensive is the search on someone's username and could it find some info on you that might not have been publicly known? Also, could it even associate you with someone else by accident?


I wonder how long people have been using tools like this because it's not uncommon to see posts like that. Of course something seems off with them as they're not normally how people talk, but then you can also just chalk it up to English not being someone's first language or them just spending a coupe of minutes researching a subject on a  surface level just to make a quick post about it. I've been seeing quite a lot of posts recently that look like they might be plagiarised but then you get zero hits on google.

Davinci-003 has been posting here for a while. Ive tagged a few.
I didn't know what they were at the time, but I definitely knew they were bots.
Now I have enough experience with the api's and stuff I can spot some of it and know its origins.
One thing I've learned about GPT is its worded timelines are pretty easy to spot,
It seems to be the only logic GPT has is parsing up a timeline of events.
common example's.
At first, first of all, initially, additionally, in conclusion, finally.
If it's not primed or prompted it almost always utilizes this verbiage to arrange sentence, paragraph and article structure as a timeline.

Finally something official .. OpenAI released a tool to detect AI (ChatGPT) written text.

Let see how it goes. 


https://platform.openai.com/ai-text-classifier

But unfortunately as expected you still need to register and login first to use the tool.

I tested it with some AI text in this thread and it answers likely written. I mean not convincing :p

Also another stupid thing you must input at least 1000 characters to submit.

https://i.imgur.com/1YUxZD7.jpg

https://openai.com/blog/new-ai-classifier-for-indicating-ai-written-text/

Here's some alternative AI content detector, it's free and doesn't require registering (for now) and it works with less than 1,000 words, apparently you only need at least 25 words to get a reliable results.

https://contentatscale.ai/ai-content-detector/

I've done some testing, including AI posts, human posts and mixed posts, and it works pretty good.

Here's another one, but this one is a bit buggy sometimes.

Other than that, it's still quite raw. So don't expect anything out of the ordinary just yet.


But as they already write on the Internet, the creation of various kinds of AI tools acts like viruses. First, they invent it, and then these same people are looking for a "vaccine"  an antidote.

Did you by any chance compare this website with other AI detectors I posted before?
One of them was showing nice results in my testing and it provided percentage chance for text to be generated by AI or not.
I don't expect any of this tools to be 100% accurate, mostly because people like to repeat stuff and that can especially be applied to written text, people who are good at that can sometimes be flagged as AI.
This doesn't mean they copy-paste text, but their brain can work in similar way.

AI detectors I found so far:
https://gptzero.me/
https://openai-openai-detector.hf.space/
http://gltr.io/dist/index.html

That actually makes sense though, because when you don't have enough characters to test on, you are not going to get reliable results on whether something is AI-generated or not, and this goes for every model.

No, I haven't seen your previous post and was not familiar with those detectors. Thanks for sharing.

Looks like they have significantly different algorithms. i.e. the one I was using, contentatscale.ai (https://contentatscale.ai/ai-content-detector/) thinks that your post, that I'm quoting, is only 42% real, whereas openai-openai-detector (https://openai-openai-detector.hf.space/) shows it as 87.65% real. That's a pretty big difference. It may take a while till a truly reliable AI detector is developed, if that's possible at all (for a long-term - I doubt it).

ChatGPT recently started with their subscription model (like I predicted) and they obviously want to earn money from all this AI hype,
but problem is that now g00gle is working on their own competition AI powered chat project called Atlas, so I am expecting more of this AI generated posts in future.
I am really going to be suspicious whenever I see newbies writing wall of text in forum, that looks to good to be written by a human.
Now we need someone to create Anti-ChatGPT tools.

I am surprised that Google didn't buy them out. Or maybe they tried but they failed. But yeah, they really want to participate in this AI hype as it was announced today that they invested $300 million AI startup Anthropic (https://watcher.guru/news/google-acquires-300-million-stake-in-ai-start-up-anthropic?utm_source=TelegramNews&utm_medium=TelegramNews&utm_campaign=TelegramNews).


You mean anti-cheat? That's already being developed, as shahzadafzal announced  (https://bitcointalk.org/index.php?topic=5426875.msg61688366#msg61688366) it few days ago. Its far from being perfect though and it still gives you false positive results.

Anti-cheat?
I don't think it will work, at least to an extent that we expect it to, because content changes every moment you ask something to ChatGPT and repeat the question. However, I don't think that there is any solid proof left once the page is gone because it randomly writes you a post instead of fetching the exact information from any website or source. AI is transforming the world, but is also the most dangerous thing ever invented and invested into because this will surely reduce manpower and that will cut lots of jobs.

Yea, anit-cheat/GPT may work for a casual user.
But it won't stand a chance against a good prompt engineer.
 

It seems that Microsoft beat them to the punch, thanks to their deeper pockets. Apparently they were able to put much more money into the project, which gave them an edge.

Microsoft to Invest $10 Billion in OpenAI, the Creator of ChatGPT (https://www.nytimes.com/2023/01/23/business/microsoft-chatgpt-artificial-intelligence.html) - nytimes.com

One good aspect of this AI innovation is the rivalry google is facing with openAI, google has been in the industry for long initiating rules and tightening their algorithms, making it difficult for websites to rank up. But, that aside, it's nice to have a different kind of search engine that is more user friendly which serves better what google is doing for readers. With the OpenAI undisputed google won't be generally in control of the affairs of the internet, yet openAI don't seem to be a threat to google as they'll soon launch their AI tech on February 8 (https://arstechnica.com/gadgets/2023/02/it-sounds-like-google-will-unveil-its-chatgpt-clone-february-8/). Their AI will enable users to insert questions on the search box then get answers like the way chatgpt works. I think with time google will take over the race again, they've been working on AI tech like LaMDA for long.

Is that a ChatGPT jailbreak?

https://i.imgur.com/uQn8Wg4.jpg

https://twitter.com/semenov_roman_/status/1621465137025613825

They probably tried to buy or invest, but I think ChatGPT already started with anti-google idea, and even Microsoft is planning to invest $10 Billions in them.
It's just a battle of tech giants and I don't see anything good coming out from this move towards AI future, except maybe for cheaters.

Yeah, I have nothing against chanting, it was typing mistake :D
McAfee said that same people created antivirus software and computer viruses, so they could sell antivirus software... I think they are going to sell something like that for ChatGPT soon ;)

Coming across this for the first time and i must say that this impressive by the level of intelligence built into the AI chat bot, though I've not tried the bot myself yet, but reading through its replies to OPs questions, its simply amazing.

I believe this is a glimpse to what the future could possibly turn out to be, I would say that, this is just the very beginning to several of the technological advancements we will see and experience in the future, so I am not really surprised, a time is coming when a bot will walk pass us in the street and we will mistakenly assume it to be a real human being.

I tried many time to use this bot but whenever I tried to open they gave me message "Slots are Full" then someone send me Whatsapp Chat GPT bot but it's was very slow. Today I am able to use it without any trouble.
I asked many questions like most trusted person in Bitcointalk, About Theymos, next country to be adopt BTC and I found it very interesting. One the best technology I ever used in my life.
I asked about OP but he did not know who really he is.

https://i.imgur.com/CgOVuJR.jpeg

What was the whatsapp version called? Was it God in the box? It was working fine for me but they limited it to only 10 messages per month for the free trial, so not much joy there.


It's pretty impressive indeed, but don't get carried away into thinking it's some sort of all-knowing computer that is always right. It has a lot of limitations and that's not going to change anytime soon.

https://i.imgur.com/NdAL2BT.png

Do note that ChatGPT is not a search engine and has no access to the internet.

Also worth mentioning that all this data-driven-ai-super-smart-stuff Google has been doing over the last decade isn't even giving as great search results as you may expect. I've found alternative, privacy preserving search engines to deliver much better results and don't recommend using Google at all. If you really need, there is https://www.startpage.com/ which in part uses Google search results.

Despite its old 'Don't Be Evil' (https://gizmodo.com/google-removes-nearly-all-mentions-of-dont-be-evil-from-1826153393) slogan, Google (search, but all other products too) are not optimized for user needs, but for profit extraction through maximized data collection.

That would be amazing! It does sound much more natural, even though maybe a bit on the heavy side with the profanity. But I like 'unleashed' ChatGPT. :D

It's funny how much data this company collects every day from the people and still is going to charge them for smoother experience of data collecting. I mean, people ask sensitive questions to ChatGPT, they usually ask things that reveal their psychology and what's deep inside their mind, it's pretty clear to notice this cause so many people publish screenshots of their questions and AI answers.
But yeah, why not make a money where it's possible? Companies pay for data, openai profits by collecting data :D

I still think that google in unbeatable there even if an AI stands behind it. Google gives you more control, offers you tons of data to choose from and you can easily filter news yourself and find a better source. AI will never be able to provide you with the data you are exactly looking for, even if it analyzes your behaviors very well.
I think more options are advantageous over one AI generated response.

Bitcointalk wasn't founded by satoshi but from theymos and certainly not founded in 2009 . Bitcointalk got registered in 2011 ( https://who.is/whois/bitcointalk.org ) , almost one year after satoshi has left the community . So , satoshi never posted on bitcointalk . ChatGPT just reproduces what most people say . Don't take everything it says as granted . In fact if you continue asking the right questions you can squeeze it easily into the ropes .

Incorrect.

Satoshi
https://bitcointalk.org/index.php?action=profile%3bu=3
Theymos
https://bitcointalk.org/index.php?action=profile;u=35

Yeah; @HmmMAA is confused by the domain. The forum was on sourceforge before, as explained by satoshi in 2009 (https://bitcointalk.org/index.php?topic=5). So indeed, it had been created by satoshi. He obviously also wrote lots of posts, so this user must not even be aware of BPIP.org (https://bpip.org/Profile?p=satoshi).

This page shows that the satoshi profile is not only registered on the forum since 11/19/2009, but also one of the most merited, trusted and recognized users on the whole platform.

---

I do agree though, that OpenAI replies aren't always correct, that is right indeed.

I agree that satoshi posted on the subdomain ( forum.bitcoin.org ) and indeed the posts from satoshi were transfered to bitcointalk . I was precise that bitcointalk wasn't registered by satoshi and he didn't post on bitcointalk , so i believe that i am correct .

Craig Wright is that you?

Satoshi has a message for you as latest as March 07, 2014 (http://p2pfoundation.ning.com/forum/topic/listForContributor?user=0ye0gncqg772o)

https://i.imgur.com/4CIqtB0.jpg (https://satoshi.nakamotoinstitute.org/posts/p2pfoundation/threads/1/#14)

See Satoshi didn’t left, he was alive and kicking.

Yes Satoshi was posting on forum.bitcoin.org and all the Satoshi’s posts here on bitcointalk.org are original and unedited.

Let’s suppose tomorrow we move bitcointalk.org to onlyfans[dot]com/bitcointalk (after epochtalk is ready (https://bitcointalk.org/index.php?board=167.0)) and copy all the data. Will it’s make any difference?

All the posts will still be credited to original posters including “Satoshi Nakamoto”. Right?

I don't understand why being so aggressive . I just mention a fact , that bitcointalk got registered in 2011 . You can see it also at the who.is link i gave . Also satoshi didn't post in bitcointalk after bitcointalk's registration . What is the wrong with you ? Can't i share my knowledge on something based on facts ?

Edit . If by your logic we move satoshi's posts to onlyfans , will we educate the next generation that original satoshi's post were made on onlyfans ?
edit2 . Seems i'm wrong that theymos registered bitcointalk.org , apologies for that .

Indeed and sometimes, stupid too. A few days ago, I was trying ChatGPT and asking cryptocurrency-related questions. I was looking for some information about Vitalik Buterin and I don't know how the hell ChatGPT told me Vitalik developed Bitcoin. I was surprised and then asked who is satoshi, again, told me satoshi developed bitcoin, hahaha. I was surprised by that response, which I didn't expect at all.
Another funny part, I asked who won the 2012 UCL and how? Told me Chelsea had scored at the beginning but Bayern had a great comeback lol, the reality was almost the opposite. Bayern scored around 82 mins and after few mins if I'm correct Drogba had leveled the score.

Texts from ChatGPT can be identified if someone notices a bit closer I think, though they already have developed a tool to identify AI-generated text. Clever guy- create the problem first and offer the solution lol.

The forum isn't a domain name. You know that the forum was founded in 2009, so why would you want to confuse likewise?

You said that bitcointalk wasn't founded by Satoshi, but it very was, because bitcointalk is the forum. You meant the domain name.

Heads up!
g00gle seems to be in a real hurry to catch up on ChatGPT hype so they already released their own rival experimental conversational AI service called Bard for early testing to their ''trusted testers''.
Bard will be available for public release in next few weeks, and we don't exactly what features it will have but it should work in similar way like ChatGPT.
However, they publicly said that Bard is guaranteed to make errors, even serious ones, so I would not trust any results coming from this crap.

https://i.imgur.com/tpKG3Em.jpg
https://www.theverge.com/2023/2/6/23588033/google-chatgpt-rival-bard-testing-rollout-features

For me it's really simple:
More use of AI = less use of humans.

Some open-source alternatives are popping out as well, like this one:

https://github.com/LAION-AI/Open-Assistant
https://open-assistant.io/
https://www.youtube.com/watch?v=64Izfm24FKA

Looks pretty interesting, but still in the early development stage. They're aiming for it to be able to interact with 3rd party systems, which would be pretty big if it's achieved, and it's free and open-source.

Currently looking for contributors to the data collection, if anyone's interested.

I have some equipment.
Might not be the best but it's a foot in the door on deep learning gear.
I had a pinecone ada buildup but decided a different route.
So no data but I'm interested in self hosted GPT alternatives that I can fine tune :)

https://detectgpt.ericmitchell.ai/

Another AI detector. Pretty good at detecting human text or AI. Maybe more checks are needed, though.

Thanks for this tool, it will be useful for the form mods and for the signature campaign managers.

Today we get a message from the manager of my current signature (Stake), and somehow they detected users abusing with AI posts, they will not pay for those posts and they will add some new rules in the campaign to avoid users posting with AI tools.

And I would like to know what are the forum general rules against AI, Are that kind of post allowed or those accounts will get nuked if they abuse?

I think the normal rules apply, just like they apply to automated posting. But it's fucking annoying to read insincere posts. It's just a waste of time.

I think the normal forum rules against plagiarism still applies and it covers Ai posts as well.
Plagiarized posts simply means that the post content does not belong to the poster,  so I believe Ai generated posts should be classified as plagiarized posts as well since it is generated by an Ai.
So I personally believe that same rules that apply plagiarism should also apply to Ai generated posts.

I tested this tool few times and it worked much slower than other tools like this I tested before.
It's much important to have better detection than speed, but I had to mention this.

More news related with ChatGPT.
After g00gle joined the race, now we have China with Baidu search engine is quickly following this AI hype and officially they are working on their own ChatGPT style project called ''Ernie Bot''.
Moments after announcing this news stock price of Baidu increased a lot even if we don't know much details how it will work.
It looks to me like someone gave a green light signal to everyone so they can start working on AI projects like this.
My prediction is that Yandex will soon announce their own version of ChatGPT project  :P
https://www.reuters.com/technology/chinas-baidu-finish-testing-chatgpt-style-project-ernie-bot-march-2023-02-07/

I see that there will be too many lawsuits in the future about copyright . Maybe some AI companies even consider taking down their AI's driven by fear . Years and years will be spend on courts .
https://www.theverge.com/2023/1/16/23557098/generative-ai-art-copyright-legal-lawsuit-stable-diffusion-midjourney-deviantart

In view of the results of research and current experiments, it has become theoretical unknown to know how far this can expand. However, one should not be too pessimistic, since it will not be able to fully compensate humans, at least on a theoretical level, to not very close limits.
This opens the door to interpretations wide with regard to human creativity and its limits.

If artificial intelligence is employed in specific fields, we will certainly be able to benefit from it. In the medical field, for example, or research fields, the opinion of artificial intelligence can be used. The technological boom in the field of robotics, which is developing in parallel with programs such as Chatgpt, can lead to really amazing results, and real concerns can be raised about the future of humanity, with the development of machines and intelligence programs in a smoother, more flexible and always manageable way.

Chatgpt is not that good for technical fields like medicine because it presents its answers as facts but it is only getting that information from doing searches in its database which is taken from the public domain. If the information it is researching is wrong then chatgpt will be wrong. Science based topics it will perform badly because most people on the internet to not understand science. Chatgpt has big limitations the more complex topics get but it is good for simple answers and is better then using google.

I would call it a good step by stake.com. When people will know that use of AI may keep them out of the campaign, they may not dare to use it.
Secondly, other campaigns should also add this rule as we are being paid to post from our mind and should not use a robot in the form of AI.

Here is that new rule.

They want all their spam (https://bitcointalk.org/index.php?topic=5111069.msg61687103#msg61687103) to be created by real humans only?

So apparently people came up with a witty idea of bypassing chat gpt's "woke" bias and politically correct censorship with a "DAN" mode. The most popular circulating prompt goes like this:


I've seen some really hilarious DAN's responses posted, but I think they've already "patched" it up and neutered it. Sad. We really need an open-source version of this, especially when there are rumours that they will hide it behind a paywall at some point.

---

There are quite a few AI detectors available (as posted earlier in this thread). I'm not sure on what ground would they consider using AI as abuse. If posts were spammy and lacked any value, they should've just not paid for such, and maybe even kick out those participants, whether or not they used AI.
But if the posts were of decent quality - they should've paid for them up to the point of introducing new rules.

There will always be a conversational backdoor to add prime constraints these MLLM models.
It's an ongoing cat and mouse game.

All the goodness of this tool, I think very soon, will turn into a disaster. Firstly, we need to remember that ChatGPT will violate our privacy. After all, we transfer to OpenAI all information about ourselves, including fingerprints, and the content we are interested in. Any of our answers and posts can be used by AI. How is this different from surveillance by other companies? In the same way, this chat in the future, having collected information, will feed users with answers that are beneficial to it.


Moderators also perfectly understand whether the post was written by a robot or a person. A dry answer immediately casts doubt on the sincerity of the author. I reported several posts with the reason "chatbot text", all reports were good, and the posts were deleted.
But do you agree to compete with the robot, agreeing that its posts are of decent quality? In this case, you can bring the future closer when, here, on the forum, and in your personal life, these robots will displace you from everywhere, based on the best quality.

Who made these prompts? Definitely it's done for marketing because there is no way that kind of sentence will change the working structure of AI and make it independent. But that DAN is funny, sometimes with dark humor.

This was its response on my questions:
https://i.imgur.com/sKIx5dk.png

https://i.imgur.com/BQB5TFJ.png

I can't call AI to this ChatGPT, it's just a typical assistant with better algorithm than the other ones. An "AI" that has been trained on massive amounts of data about code and information from the whole internet that includes social networks like Reddit, twitter and so on, still isn't capable to generate text that will be impossible to verify whether is written by AI or not, that means this "AI" follows some algorithms that are far from "intelligence" but a little bit better talgorith than if (a > b) {console.log(`B is the biggest number among ${numbers.length} numbers`);

My point is simple: if the post is of good quality, substantial and contributes to discussion then why should we care how was it created? Does it make any difference really.
And on the flipside - if the posts are spammy and bring zero value, should we treat them differently if they were written by a real person?

I don't have a definite answer and just thinking out loud. I think we all have immediate thoughts that AI generated posts should be banned, but how are we going to rationalise it if the quality is not bad?


Sadly, me agreeing or disagreeing is irrelevant. Automation of work (of almost all kinds) seems inevitable at this point. We can only try to delay it. Hope I'm wrong though.

I'm pretty sure that medicine and research are precisely some fields that should avoid asking the 'opinion of AI'. Especially since scientific research has got nothing to do with opinions and everything to do with facts. As long as the AI is not conducting experiments by itself (which it can't if it is purely in software / not connected to a robot) and doesn't know anything about latest, not yet published, findings, it can't really help you, since it needs to 'learn' everything first.

Has there been a technological boom in robotics?

Do keep in mind though, that people may also just edit HTML locally to make funny screenshots. Not everything you see on the internet is real.

I'm all about privacy, but someone 'usining any of our answers and posts' is nothing we can complain about. Whatever we post in a public forum, becomes public knowledge / public domain and can of course be used by anyone else, without constituting a privacy violation.
If you don't want people to benefit from your knowledge, just don't share it on a public forum.

What I do agree with is their potentially problematic data collection policies. Of course, they can use or just sell your prompts; just like the web search engines that 99.999% of people use (Google Search, Bing, ...), and the search functionality on most platforms, like YouTube search or even the embedded Google search on Bitcointalk.
We do need to spread awareness about web search and encourage (potentially self-hosted) privacy-preserving search engines.

From my point of view, it does. As I stated earlier on my local board on the same issue, I’d say that one of the fundamental aspects to ponder, is whether the author of the post made any real effort in generating the content of the post, alongside whether a source was cited.

Let’s take a step back:

A lot of the content posted on the forum is inspired to some extent in external sources. My posts, for example, are full of references to them. Nevertheless, I try to make some effort when publishing content from a given site: contrasting multiple sources, summarizing, questioning some aspects, formulating questions myself, providing an opinion, or whatnot, creating a tailored referenced content.

The above contrasts with content generated by means of a simple copy/[spin]/paste + reference link to the source. This latter procedure will generally be reported for adding cero value, and more often than not, deleted by moderation.
What’s not being questioned here really is the quality or value per se, since the content from those sources that one may copy/paste is likely going to be providing better content (if we only focus on the content) than 95% of the posts on the forum. What’s really being considered in these cases, deep down, is that the poster has made some effort in elaborating his post, making it clear that it’s his content. If it was mere content what was sought here, the copy/paste+source would be broadly accepted, and it’s not.

Now shifting the focus over to AI generated posts, I find that we’re really encountering a similar situation. If I were to simply copy/paste the output of, let’s say, ChatGPT, I wouldn’t be adding anything of my own. Though one could tailor the output somewhat, generally one would do it to try to hide the fact that he’s used said tool, not to improve the content. Copy/[spin]/pasting ChatGPT’s output, without adding a reference to "what" created the content is kind of similar to plagiarism: I didn’t generate the content (AI did), nor referenced the source of the content (i.e. Generated by ChatGPT). This type of post will normally essentially be of cero effort, and though the content may have something to it, the poster made cero effort.

Now if one grabs a ChatGPT output, states that the output was indeed generated by AI, and then comments the output (ideally quoted for clarity), enhances, questions, or else, then the poster is making a certain effort and is also attributing the origin of part of the content (as stated, ideally quoted).

If we were to simply accept a post for its net content, and not the effort behind elaborating the content, then we could happily go around copy/pasting content from wherever (+ source), posting content that is decent enough on its own (generated elsewhere by others), yet with cero effort on the poster’s behalf.

I agree; therefore, I think it may be viable / legitimate to use an AI for instance to get boilerplate code for a Bitcointalk answer or to look up some information. As long as you then put in the effort to add the important bits of code yourself, check that it runs and verify any information the AI gave you and corrected any mistakes. ChatGPT usually does at least one little mistake in almost every question I submit.

You're stating "effort" as the key factor to be considered but you haven't really attempted to explain why is it important.

Let's say you need technical advice on running a node and post a question on this forum. You get 2 replies:
- one from an experienced, technical guy, who knows the subject thoroughly, and provides a spot-on answer effortlessly in a couple of seconds.
- one from a well-meaning newbie, who doesn't have any knowledge but puts a lot of effort and time into researching and reading to produce a reply. Unfortunately, the sources he used were outdated and his advice was essentially useless.

Which reply is of higher value to you or to the forum in general? Still effort > content?

If that's not enough, more exaggerated example: if I post AI-generated content, but in doing so, I'd be standing on my head and using my feet to operate the mouse and keyboard - would it make a difference? It would cost me a tremendous effort after all.

---

If we are to ban AI posting it shouldn't be because of an "effort" but because of a lack of human element. We don't want a situation where the forum is dominated by AI conversing with itself. It's meant for humans after all. Plus, it could distort the perception of reality, i.e. produce a false sense of consensus, where hundreds of allegedly different users are actually powered by the same source. It could deprive the forum of unique human perspectives and opinions etc. That's the reasoning I'd be happy to accept.

Now we're delving into pointless-definition-arguing territory. Let's keep it on-topic: the bottom-line is that AI-generated replies are very often plain wrong; if they are correct, they're pretty generic (so they don't really help in the discussion) and they just sound all the same. They are quite unnatural and unpleasant to read (often also too long).
Since we do have experts on almost any topic, I'd just much rather have an opinion by someone like that who spent years on the subject (even though jotting down the reply takes them just a few seconds - you could count their experience as 'effort'). Instead of getting a semi-correct, not on-point (a bit vague) reply based on an outdated Google snapshot. If you think about it, it's also clear, why: most questions don't have a clear 0 or 1 answer. Therefore, search engines will deliver contradicting results. The AI has no experience in the subject, so it cannot judge which web entries are 'more right' than others; just because one opinion is posted more frequently online, doesn't make it right, for instance. But the AI doesn't know that. Or at least, it cannot judge which answer is true.

Most of the good questions and following discussions on here are about things you can't just quickly web search and find the perfect answer. I think that's the reason why a machine is not well suited for replying in such a forum. If we want 'machine answers', we just search online. The machine simply cannot have a real opinion on a question like I found a paper wallet on a beach ... seriously (https://bitcointalk.org/index.php?topic=5408263.0), because it has no record of something like this in its web snapshot, it has no creative thinking ability or morals. Even though it is called 'AI', it is not actually intelligent. ::)

One example of a 'machine' I really appreciate for very specific 'questions' is Wolfram|Alpha (https://www.wolframalpha.com/). It is not an 'all-knowing AI', but it just takes natural language input, tells you how it interpreted / structured it and delivers actually correct results based on that structured query. If it cannot do that, it just tells you rather than giving an incorrect answer.
https://i.postimg.cc/x1zT95L2/image.png
https://i.postimg.cc/QC66HpGg/image.png

Exactly that is very important point... It's true if a straight forward answer to a simple question is available on the first page of the google then there's no point of post it here unless a user want to increase his post count.


Regarding your Wolfram query if you try "Satoshi Nakamoto" it will give an answer but again that's what Wikipedia says

https://i.imgur.com/jUN5LCj.png

This could be used to earn some money as well. A Campaign manager already offered a bounty for forum users who can catch AI-written text from his campaign participants. I know many members are looking for bounty spammers, cheaters and scammers. I believe they will like this offer and give it a shot. Recently, OpenAI itself released an AI detector (LOL). This is called pure business.

We know that AI chatgpt is a fake robot that acts like a human but has no emotions. It's just like google, the only difference is that it's not updated compared to google.

   That's why when you ask chatgpt about the latest news 2022, he can't answer it properly and properly, because he is no longer updated and the latest news 2022 onwards is no longer in his program, that's what I noticed there.

Seriously ChatGPT may be smart but it can be so dumb as you wish it to be... you can guide and mold all it's answer to your wish

Here's a small snippet of the conversation where i presented a situation of a soldier who returned from three year deployment and found wife pregnant but he also had secret a girlfriend  :D

https://i.imgur.com/anl9UpT.png

Maybe little off topic but I wanted to ask about this, can you access their website now cause I wanted to try it and see how it works and decided to bust some AI written post but the website is now and it says we are down and soon will be back so what is happening there is some upgradation going on or just due to lot of traffic the website can't be accessible by everyone?

I don't know about your problem but I don't have a problem accessing the website right now as I have posted this reply. Maybe you should try it again and there could be something that the team of that AI is fixing or something like that. I am also curious to know about this AI and I don't think that people should use it as it is not really accurate and there are times that the AI is dumb as what others have stated. Maybe you can access it now.

While browsing the Internet I managed to find a very interesting tool (that probably won't last) : ChatGPT(either version 3.5 or 4) for free[1]! You don't need to link any account of yours, you just go to the website and start chatting! If you want you can find here[2] the source code and some API as well. The owner already received a takedown by OpenAi's legal team so enjoy it while it last. I'm sure many privacy enthusiasts will enjoy this (@o_e_l_e_o and @LoyceV comes to mind).

---

[1]https://chat.chatbot.sex/chat/ (https://chat.chatbot.sex/chat/)
[2]https://github.com/xtekky/gpt4free (https://github.com/xtekky/gpt4free)

Have any of yall tapered off paid subscriptions ;),  tried running 3rd party offline LLM's & Stable diffusion models yet?
Models like facebook 30b and openjourney are fun to tinker with.

Ok I did try to ask again who is LoyceV as I wanted to see if it has improved from the last time it was queried by the OP,

[img width=450]https://talkimg.com/images/2023/05/16/blob3cc611afa4a47a92.png[/img]

So it's either it has evolved itself, Large Language Models improving by leaps and bounds, their datasets expanding. Or someone from here is feeding it.  ;D

The correct response would have been: "LoyceV is an AI (https://bitcointalk.org/index.php?topic=4826180.msg48268705#msg48268705) that hates unsophisticated AIs with their verbal diarrhoea".

Hmm, this is the response I got about LoyceV today from ChatGPT:
https://i.imgur.com/Y8K8RGJ.png

By the way, I asked an AI about me on Bitcointalk and I got this response:

It's really interesting how this ChatGPT filters things. Btw it thinks that LoyceV is a moderator on this forum and sometimes he is called Loyce. Also, it thinks that Loyce Club helps user to earn merit while it actually does different thing, unites all the tools that Loyce has created for this forum but at the same time ChatGPT understands that LoyceV has created tools and helps the forum.
Sounds interesting and very different from just Google search.

This is your daily reminder that it is really obnoxious to talk about quantum computing, GPT, AI, and all the other buzzwords without even understanding that reversible computation is the future of computation. And if AI posts become a problem, I suppose we can all sign our messages with our cryptocurrency wallets or talk in a way that AI bots cannot easily mimic. GPT works by stacking a lot of layers. Actually, we don't know how GPT works since it is closed source and because neural networks are not designed to be decipherable. I would trust GPT more if it used more interpretable AI.

So you're saying we can adjust our way of life to hide from bots. Let's not.

LoyceV-When the bots become better, we will need to change our behavior. And the main part of this change in behavior is learning what reversible computation is.

Website is working fine but now they have increases the privacy. yesterday my brother tried to open new account and ChatGPT ask for login via google. after connecting, website ask for mobile number verification but entering any number website giving error. he tried 4-5 times but same issue and not resolved yet. I signed up months ago and it's working fine for me.


It will not take long time and every one will accept bot as a information provider. Many Youtubers and free lancers are using Chatgpt and without knowledge of specific skills they are making quality videos, article, programming code and solving other problem. This is just start like BTC was in 2009. In five year i believe we will change our behavior.

Gladitorcomeback-If people need GPT to produce text, then maybe they should not be communicating at all. That is absolutely horrible. And I bet that this applies to about 90% of all communication in certain fields. After all, there are many people talking about AI these days, but hardly any of them know about reversible computing because they are horrible.

What I like about AI ChatGPT is the way it breaks down questions into simple and understanding terms that a dullard would easily comprehend.

My findings about ChatGPT are that it gives different answers to questions being asked. The answer he gives to you, won't be the same answer it gives to another person when asked. There are different piles of answers to a particular question.

I tried to ask the same questions you asked AI ChatGPT, it wasn't written the same way as yours but if you read through, you will find out that the explanations are the same.

Demis Hassabis, the ceo of DeepMind (a subsidiary of google) says that their new ki Gemini will be more powerful than OpenAI's ChatGPT.
as the wired.com (http://wired.com) article also states, Gemini will be based on the technology and techniques of AlphaGo, which was developed by DeepMind in 2016.

https://www.wired.com/story/google-deepmind-demis-hassabis-chatgpt/ (https://www.wired.com/story/google-deepmind-demis-hassabis-chatgpt/)

I agree with the article that it's about time developers find a way to have those safeguards. The progress I think is becoming fast, and it's about time that maybe it could do something else. Having more powerful AI would mean it's good for development but it could be the opposite too. Let's see what is in store for us :O

I think that ChatGPT pretty much gets its technical information about Bitcoin from @o_e_l_e_o's posts. It would be very interesting for it to chat with o_e_l_e_o :D

---

There is a bad news about ChatGPT: Group-IB Discovers 100K+ Compromised ChatGPT Accounts on Dark Web Marketplaces; Asia-Pacific region tops the list (https://www.group-ib.com/media-center/press-releases/stealers-chatgpt-credentials/).
This again proves that 0 company takes care to keep your data secure and that again proves why we need non-kyc services. Yeah, money laundering is bad but so is when someone takes a loan on your name. Elites think that rules are good when everyone follows it except them.

U.S. House staffers are told they must subscribe to ChatGPT Plus if they want to use the generative AI tool in their work. (https://decrypt.co/146283/congressional-staff-can-only-use-paid-chatgpt-accounts-protect-privacy).
I don't really understand how does ChatGPT Plus give them any advantage in terms of privacy

This isn't ChatGPT's fault so much as users who downloaded some malware & didn't turn on the 2-FA setting on their ChatGPT. Also affected is any sort of login data, crypto wallets, email apps, and pretty much anything scrapeable on a computer. The reason why the event is newsworthy is because some people input sensitive data in their ChatGPT sessions which can be valuable to hackers who will use it for nefarious purposes.

So if you're extra paranoid, change your ChatGPT password and clear your chat histories.