Minichan

Topic: Learning anything practical on the internet sucks now

Anonymous A started this discussion 2 years ago #113,737

Used to be you could find small websites dedicated to just about anything. They written from anyone ranging from a hobbyist to someone who obsessively knew the subject. You could easily find blogs, guides, personal anecdotes, and pictures written by these people.

Now when I search for anything I can only find shitty generic "top 10 tips" with stock images, or a useless article that is little more than an advertisement for a related service or product (often only servicing a small geographic region and thus being completely irrelevant to probably 95% of people that come across the page).

The closest I can still find to the small websites of the 90s-00s is via youtube videos. And even many of those are gimmicky and from a contrived angle of an "expert" informing all of us drooling retards, rather than a more honest, straightforward "here's what I know, here's what I did, here's how it turned out" that I'm looking for. Also, videos suck for learning and are no replacement for a well-written guide, article or blog post with pictures.

Hitler was right. The internet was a mistake.

Anonymous B joined in and replied with this 2 years ago, 8 hours later[^] [v] #1,259,440

Agreed. Thank the scummy SEO guys who've succeeded in poisoning every search engine.

I still find random people's guides on Reddit to be useful. Some subreddits are filled with passionate hobbyists who'll spend their time helping you and writing nerd guides and how-tos. It's not the same, but it's something.

Killer Lettuce🌹 !HonkUK.BIE joined in and replied with this 2 years ago, 33 minutes later, 9 hours after the original post[^] [v] #1,259,441

@OP
Blogs like that are still out there. I probably dodge a lot of the clickbait stuff you're describing because a lot of my searches are about niche history stuff, but I regularly find plenty of personal blogs with interesting information.

@previous (B)
Yeah, I was about to mention Reddit myself. Usually even if you have a very niche interest, you can find a Subreddit dedicated to it, or at least a post that can help you out.

(Edited 3 minutes later.)

Anonymous A (OP) replied with this 2 years ago, 8 hours later, 18 hours after the original post[^] [v] #1,259,461

@1,259,440 (B)
@previous (Killer Lettuce🌹 !HonkUK.BIE)
Yeah, it seems all the actual people migrated to either reddit or youtube, while the rest of the internet was taken over by SEO shit and now, probably, GPT-written articles. The whole thing of adding "reddit" or "site:reddit.com" to my search queries is definitely something I've had success with, although it depends on the topic.

It's really a shame everyone had to migrate to commercial, centralised platforms and away from personal websites and independently run forums. This destroyed some of the main benefits of the internet to begin with.

Anonymous D joined in and replied with this 2 years ago, 2 hours later, 20 hours after the original post[^] [v] #1,259,476

The dead internet theory strikes again

!MLHqI35Srs joined in and replied with this 2 years ago, 10 hours later, 1 day after the original post[^] [v] #1,259,554

Nobody will need to know anything. The way being artful with any of those things had died like the S. Xiatyl futurism films whereby all that was left was papyrically-empty literature and a metiziac square-plattafice temple.

I personally don't mind WYSIWYG freeware bloghosts to substitute HTML knowledge and people have continuously been telling me to fuck off from programming because people only want homoerotic gimmic-games involving no statisticometry and no specialist knowledge of political-economy.

EDIT: Fucked completely on Mulberry-liquids in prison, I had laughed over and over and over at that one trope being applicable to the west. So uncontrollable was my laughter I was banging my head to stop myserlf. It's' what convinved me to drop National Socialis, for Juche sasang and (re-)learn Korean, which I did speedily under its' influence.

(Edited 3 minutes later.)

Anonymous A (OP) replied with this 2 years ago, 9 minutes later, 1 day after the original post[^] [v] #1,259,556

@previous (!MLHqI35Srs)
> homoerotic gimmic-games involving no statisticometry and no specialist knowledge of political-economy.
I never play anything else.

!MLHqI35Srs replied with this 2 years ago, 3 minutes later, 1 day after the original post[^] [v] #1,259,558

@previous (A)
With retard bans capping my finances and even Facebook banned as a gaming platform (which I've decided to respect; its' them vastly misinterpreting their Modus Operandi) I can no longer recall the name, but the most gimmicy game I've played is a Sim City-spinoff for the site.

I suppose Gran Turismae count as gimmic games. I'm guilty of that as of 5 years' ago (I bought a PlayStation 3, owned it for 3 years', then wanted rid of the thing; its' either in storage or police reclaimed since and I haven't bothered the fuck checking ever since forced to leave the pseuderlatively named 'home-town'), reclining off the internet by state force of imposing an order the direct consequence of speaking to LagoonaBlue and her negro Glaswegian transvestite counsellor friend.

(Edited 1 minute later.)

Anonymous B replied with this 2 years ago, 46 minutes later, 1 day after the original post[^] [v] #1,259,566

It's about to get worse:

AI content is no longer inherently bad

Before this update, the Google Guidelines explicitly stated that good content should be created by people, but this limitation has now been removed. Google has apparently realised that they cannot win the battle against AI content and is now focusing on the quality of the content, regardless of its origin.

https://www.sistrix.com/blog/google-helpful-content-update-september-2023/

Anonymous A (OP) replied with this 2 years ago, 28 minutes later, 1 day after the original post[^] [v] #1,259,574

@previous (B)
They had no way to stop it. They really should've gone the other way and leaned in and pushed more on developing large language models. They should've had some type of integrated search GPT-like assistant well before OpenAI and Bing. Google is a search company for christ sake. Some washed up 90s software company shouldn't be the first to make changes to the search game.

As for the AI-SEO-shit-pages, it'll definitely get a lot worse. But for some topics, likely those either more technical or, ironically, far less technical (like subjective skills such as writing, or art), I think AI articles could eventually be better than human-created sites. For that, though, the training of them is going to need to move quite far beyond simply replicating/predicting human text. It's going to need to move beyond the Kimmo stylesheet altogether. It can and will be done with enough time and research; it's just not as simple as throwing more GPUs at the problem like it seems all OpenAI is capable of doing these days.

I look forward to reading about how to understand human literature, art, history, or some technical subject from the perspective of a super intelligent agent more capable than any human that has ever existed. I think there's quite a substantial chance that will happen in both of our lifetimes.

Anonymous B replied with this 2 years ago, 5 minutes later, 1 day after the original post[^] [v] #1,259,576

@previous (A)
> I think AI articles could eventually be better than human-created sites. For that, though, the training of them is going to need to move quite far beyond simply replicating/predicting human text. It's going to need to move beyond the Kimmo stylesheet altogether.

Yep, that's probably right. There's no reason to think the pace of AI will slow, so it'll almost certainly produce better-than-human content eventually. The question (especially as it relates to the "decline" of good Internet content) is when? I hope we don't see another decade of dreadful, spammy bot content before we get decent articles etc.

Also, I personally want to know when content is AI-authored. It should never be undisclosed or disguised. I think it's important to know when content is human or bot.

(Edited 8 minutes later.)

Anonymous B double-posted this 2 years ago, 2 minutes later, 1 day after the original post[^] [v] #1,259,578

@1,259,574 (A)
> I think there's quite a substantial chance that will happen in both of our lifetimes.

Also agreed. Consider how far AI has come in just the last decade. There's obviously still a significant way to go, but there's no reason to think it'll take 100+ years to get there (or to doubt we'll ever create intelligence at all, as some do).

Anonymous A (OP) replied with this 2 years ago, 3 minutes later, 1 day after the original post[^] [v] #1,259,579

@1,259,576 (B)
> Also, I personally want to know when content is AI-authored. It should never been undisclosed or disguised. I think it's important to know when content is human or bot.
Actually I disagree with you on that point. Not that I disagree with you that it's important to know (at least today in the current era of its development), it's just that I don't think anyone ought to have an obligation to say. Only exception to that is sufficiently large corporations. They ought to have requirements imposed on them that individuals and small groups of individuals do not. Disclosing AI content I think ought to be among them for the likes of Google, Microsoft, Apple, etc.

Anonymous A (OP) double-posted this 2 years ago, 2 minutes later, 1 day after the original post[^] [v] #1,259,580

The other issue with disclosure is that it'll become increasingly unenforceable, at least to impose on individuals and small groups of them. And making unenforceable policies and laws is just bad practice, even if you don't share some philosophical objection to the idea to begin with.

Anonymous A (OP) triple-posted this 2 years ago, 2 minutes later, 1 day after the original post[^] [v] #1,259,581

@1,259,578 (B)

> >I think there's quite a substantial chance that will happen in both of our lifetimes.
>
> Also agreed. Consider how far AI has come in just the last decade. There's obviously still a significant way to go, but there's no reason to think it'll take 100+ years to get there (or to doubt we'll ever create intelligence at all, as some do).

I'm always surprised by the number of nominally non-religious people that believe there's something about human cognition that cannot be replicated in a machine. It's crazy. We are machines. Implementing an algorithm in silicon instead of carbon. There's no reason to believe it will make any difference. And every piece of evidence we've seen thus far indicates that it will not and cannot make any real difference.

Anonymous B replied with this 2 years ago, 2 minutes later, 1 day after the original post[^] [v] #1,259,582

@1,259,579 (A)
@1,259,580 (A)
What about some sort of industry standard ethical guidelines, instead of enforcement by law? I could see that becoming a thing: a voluntary "code of conduct" that large publishers etc. could choose to adopt, to highlight content made entirely by AI or with AI assistance.

I'm sure over time, as AI grows in "intelligence" and ability, the distinction will become more and more meaningless. But until then, I think it's a decent enough idea.

Anonymous B double-posted this 2 years ago, 1 minute later, 1 day after the original post[^] [v] #1,259,583

@1,259,581 (A)
Yeah, unless you believe in the supernatural (e.g. "souls") there's no reason to think we're anything special. We might be complex, but it's all just particles interacting at the end of the day. We'll get there.

I think humanity desperately wants to hold on to the idea there's something profoundly special and supranatural about the human mind. To imagine ourselves as just mechanical parts is almost insulting.

(Edited 1 minute later.)

Killer Lettuce🌹 !HonkUK.BIE replied with this 2 years ago, 27 minutes later, 1 day after the original post[^] [v] #1,259,587

Personally I'm undecided if something artificial could every perfectly replicate the human mind or the human experience, I suppose that's something for scientists and philosohers to sort out down the line. But I think it's probable that it could get so close that it'd be a distinction without a difference anyway.

Not to be too doom and gloom, but I'm starting to feel like this is the last chance, at least for a lot of people, to do creative things like make art or write or make videos or whatever. As AI gets better and can match humans even more, and the barriers of entry will in turn go down, everyone will get using it everywhere. Already on YouTube, I'm seeing whole channels that use purely AI generated images for graphics, and it's trivially easy to get ChatGPT or whatever to write you a script or to enhance one you wrote.

!MLHqI35Srs replied with this 2 years ago, 4 minutes later, 1 day after the original post[^] [v] #1,259,588

Existentars aren't even engaging in philosophy.

S. Xiatyl morbid situationists had it at you're all papyriing.

Killer Lettuce🌹 !HonkUK.BIE replied with this 2 years ago, 5 minutes later, 1 day after the original post[^] [v] #1,259,589

@1,259,587 (Killer Lettuce🌹 !HonkUK.BIE)
I suppose the counterpoint here is that it's still creativity if you're directing an AI and then assembling its assets to make a video or a piece of art and so on. I've seen some of these AI YouTubers, and the end result is still okay. Maybe the real lesson here is to adopt AI or be left behind, which is still pretty depressing tbh, humans drawing and writing things by themselves is an impressive and valuable thing.

boof joined in and replied with this 2 years ago, 7 hours later, 1 day after the original post[^] [v] #1,259,616

@1,259,581 (A)
we don't even know how cognition arises in biological form in the first place -- but we do know that whatever current computers are doing, it is not thinking.

Anonymous A (OP) replied with this 2 years ago, 2 hours later, 1 day after the original post[^] [v] #1,259,658

@1,259,582 (B)
> What about some sort of industry standard ethical guidelines, instead of enforcement by law? I could see that becoming a thing: a voluntary "code of conduct" that large publishers etc. could choose to adopt, to highlight content made entirely by AI or with AI assistance.

Yeah, I have no objection to this. I don't think it would be followed by the people that ought to follow it the most, but if they did, how could I argue with people voluntarily doing what they choose especially if it's what I also want them to do heh?

Anonymous A (OP) double-posted this 2 years ago, 20 minutes later, 1 day after the original post[^] [v] #1,259,662

@1,259,587 (Killer Lettuce🌹 !HonkUK.BIE)

> Personally I'm undecided if something artificial could every perfectly replicate the human mind or the human experience, I suppose that's something for scientists and philosohers to sort out down the line. But I think it's probable that it could get so close that it'd be a distinction without a difference anyway.

Please elaborate. Isn't your degree in philosophy? Anyway, this point of view makes zero sense to me. And you call yourself an atheist! I thought the whole point was to reject views that seem implausible, like Santa Clause. What is plausible about the human mind being special and somehow being unbounded by the heretofore inviolable laws of physics? Rejecting that viewpoint is the natural extension of the Copernican principle and this principle has led to scientific progress for nearly 400 years now.

It's not that I think it's inconcievable and impossible that there's something special about the human mind. Maybe there is such a thing as a soul. I'm not writing the idea off entirely. But all the evidence seems to point away from that being the truth.

Hell, I write off the idea of a soul far more than the idea of a god itself. Something from nothing? How does that make sense? It doesn't. A god doesn't really solve that "problem", but if we're starting off from a premise of nonsense, I guess anything goes?

I guess that's just a long way of saying that even as an agnostic non-atheist, I think the idea of a soul is far, far sillier than that of a god.

> Not to be too doom and gloom, but I'm starting to feel like this is the last chance, at least for a lot of people, to do creative things like make art or write or make videos or whatever. As AI gets better and can match humans even more, and the barriers of entry will in turn go down, everyone will get using it everywhere. Already on YouTube, I'm seeing whole channels that use purely AI generated images for graphics, and it's trivially easy to get ChatGPT or whatever to write you a script or to enhance one you wrote.

Neither of us knows how things will turn out. But people that worry excessively about new technology I think always end up on the wrong site of history. I'm sure you can think of as many if not more examples of that happening than I can.

Maybe some of those fears will come true, but maybe some will become irrelevant in a post-consumer-producer society. If you, I, and all of us no longer need to work and produce something new or creative, who cares? I don't know about you, but a lot of what I do and learn about is useless. I'll never be able to produce widget X Y or Z at the level of someone who does this for a living, or even, in many cases a lifeless machine. Doesn't mean I can't take joy in doing things myself anyway.

For some transient amount of time I think humans + AI will achieve more than humans alone. But I think that window will close and any human intervention in the creation of a product or idea will just fuck it up. Like a small child "helping" to clean up or build something.

Nihilism is and always will be a problem. That's nothing new.

Anonymous A (OP) triple-posted this 2 years ago, 3 minutes later, 1 day after the original post[^] [v] #1,259,663

@1,259,616 (boof)

> we don't even know how cognition arises in biological form in the first place
If we can create something that is indistinguishable in every measurable way (aside from contrived measures like "oh it doesn't have cells or DNA") then what ground is there to say that it's different?

> but we do know that whatever current computers are doing, it is not thinking.
What does this even mean? What is "thinking"? Come up with a measurable objective test of it and I guarantee you AI will be there to meet and beat us at it within 2 decades max.

Anonymous A (OP) quadruple-posted this 2 years ago, 2 minutes later, 1 day after the original post[^] [v] #1,259,664

@1,259,589 (Killer Lettuce🌹 !HonkUK.BIE)

> Maybe the real lesson here is to adopt AI or be left behind
For a short amount of time, yes. In a longer amount of time, the lesson will be don't fuck up and get in the way of the AI, you'll just make things worse.

> humans drawing and writing things by themselves is an impressive and valuable thing.
Disagree. What's better is better.

boof replied with this 2 years ago, 33 minutes later, 1 day after the original post[^] [v] #1,259,670

@1,259,663 (A)
well on the grounds that is actually different. as for your peculiar desire to have a measureable objective test of thinking, why bother? We know that these devices do not think. We made them, we know how they operate.

Anonymous A (OP) replied with this 2 years ago, 35 minutes later, 1 day after the original post[^] [v] #1,259,679

@previous (boof)
So it sounds like you have a religious, untestable belief. That's fine. Just call it what it is.

boof replied with this 2 years ago, 9 hours later, 2 days after the original post[^] [v] #1,259,789

@previous (A)
no no, that was you that did that

Anonymous A (OP) replied with this 2 years ago, 2 minutes later, 2 days after the original post[^] [v] #1,259,792

@previous (boof)
No it was you.

boof replied with this 2 years ago, 1 minute later, 2 days after the original post[^] [v] #1,259,795

@previous (A)
oh no, it most certainly was not

Anonymous A (OP) replied with this 2 years ago, 2 minutes later, 2 days after the original post[^] [v] #1,259,796

@previous (boof)
Now let's get thing quite clear: you most certainly did.

Anonymous H joined in and replied with this 2 years ago, 2 hours later, 2 days after the original post[^] [v] #1,259,801

@1,259,670 (boof)

> We know that these devices do not think.

How can you say that, if you cannot define what thinking is?

Anonymous D replied with this 2 years ago, 3 hours later, 2 days after the original post[^] [v] #1,259,815

@previous (H)

> >We know that these devices do not think.
>
> How can you say that, if you cannot define what thinking is?

This. Machine Intelligence will just be an emergent property of running a bunch of neural nets.

Catherine !TGirlYJKXM joined in and replied with this 2 years ago, 3 hours later, 2 days after the original post[^] [v] #1,259,833

I'm surprised nobody created a search engine that finds exactly what you wished to search for honestly.

boof replied with this 2 years ago, 32 minutes later, 2 days after the original post[^] [v] #1,259,837

@1,259,801 (H)
I don't have to define it. It is a known word in the language.

Anonymous A (OP) replied with this 2 years ago, 2 hours later, 2 days after the original post[^] [v] #1,259,852

@1,259,833 (Catherine !TGirlYJKXM)
That's what google is supposed to be doing. But it looks like a dead 90s software company could beat them to it, as I said earlier. Google's too busy serving up shitty SEO top 10 articles while tracking our every eye movement, breath, and bowel movement of everyone on the planet.

Anonymous H replied with this 2 years ago, 6 minutes later, 2 days after the original post[^] [v] #1,259,855

@1,259,837 (boof)

> Thinking. noun. The process of considering or reasoning about something.

Please explain how the machine does not think by this definition.

boof replied with this 2 years ago, 2 hours later, 2 days after the original post[^] [v] #1,259,895

@previous (H)
there is no considering and reasoning occurring by the processing that these devices do. the processing is fully describable by the inventors of these devices. we know how they work. they don't do it with conscious thought.

Anonymous A (OP) replied with this 2 years ago, 13 minutes later, 2 days after the original post[^] [v] #1,259,899

@previous (boof)
> conscious thought
Kicking the can further down the road :)

boof replied with this 2 years ago, 4 hours later, 3 days after the original post[^] [v] #1,259,945

@previous (A)
what can, we are talking about automatons and our disagreement that they are beings of sense and sensibility

(Edited 1 minute later.)

Anonymous A (OP) replied with this 2 years ago, 11 minutes later, 3 days after the original post[^] [v] #1,259,949

@previous (boof)
The can labeled "the core of the argument" is the one you kicked down the road.

Anonymous G joined in and replied with this 2 years ago, 3 hours later, 3 days after the original post[^] [v] #1,260,004

Anonymous H replied with this 2 years ago, 5 hours later, 3 days after the original post[^] [v] #1,260,034

@1,259,895 (boof)

> they don't do it with conscious thought.

So now there are new requirements required for something to qualify as "thinking". Earlier you said that you don't need to know what thinking means because there is a dictionary definition, but when provided with the dictionary definition you started to add new requirements for it to be "real thinking".

Anonymous J joined in and replied with this 2 years ago, 5 hours later, 3 days after the original post[^] [v] #1,260,060

I realized google search results went to shit when Wikipedia articles are hard to find now

They used to always be the first or second result

boof replied with this 2 years ago, 25 minutes later, 3 days after the original post[^] [v] #1,260,063

@1,260,034 (H)

> >they don't do it with conscious thought.
>
> So now there are new requirements required for something to qualify as "thinking". Earlier you said that you don't need to know what thinking means because there is a dictionary definition, but when provided with the dictionary definition you started to add new requirements for it to be "real thinking".

you need to reread what I wrote genous

boof double-posted this 2 years ago, 1 minute later, 3 days after the original post[^] [v] #1,260,064

@1,259,949 (A)
I did a word search and your reply is only instance of that

Anonymous H replied with this 2 years ago, 2 hours later, 3 days after the original post[^] [v] #1,260,086

@1,260,063 (boof)

@1,260,063 (boof)

I read it properly the first time, you said two things that there is no reasoning or considering performed by the machines (false) then that we can perfectly describe and understand the process and it isn't done with concious thoughts (which means that you are saying there are two additional requirements for something to be thinking, that we can't describe it and it uses conciousness).

boof replied with this 2 years ago, 1 day later, 4 days after the original post[^] [v] #1,260,405

@previous (H)
that's quite the extraordinary claim you had that these computers are performing actual cognition. now you are claiming that it performs tasks that are not actually involving thinking. Well then, it turns out we were in agreement the whole time. The devices are automatons of a sort.

Anonymous H replied with this 2 years ago, 25 minutes later, 4 days after the original post[^] [v] #1,260,417

@previous (boof)

Are you smoking crack? What claims are you talking about?

boof replied with this 2 years ago, 6 hours later, 5 days after the original post[^] [v] #1,260,468

@previous (H)
dang, that's one short memory you got there chief

Anonymous H replied with this 2 years ago, 5 hours later, 5 days after the original post[^] [v] #1,260,496

@previous (boof)

You should reread my posts geoius

boof replied with this 2 years ago, 7 hours later, 5 days after the original post[^] [v] #1,260,533

@1,259,801 (H)
here is your first post. I reread it. It is still implying that you have the idea that these machines think. Later, you dissociate yourself from the idea. It would seem you agreed with me the whole time, and was merely being silly.

Anonymous H replied with this 2 years ago, 2 hours later, 5 days after the original post[^] [v] #1,260,555

@previous (boof)

> Later, you dissociate yourself from the idea

I would be highly interested to discover where I did that

Anonymous H double-posted this 2 years ago, 1 minute later, 5 days after the original post[^] [v] #1,260,556

Also

> It is still implying that you have the idea that these machines think.

No, it's implying that you cannot rule out the idea that these machines think, because you cannot clearly delineate what "thinking" means.

boof replied with this 2 years ago, 10 minutes later, 5 days after the original post[^] [v] #1,260,564

@previous (H)
why don't you pick on my lack of definition of every fucking word I used? fucking contrarian prick

boof double-posted this 2 years ago, 46 seconds later, 5 days after the original post[^] [v] #1,260,567

oh this rock doesn't think

H: oh yeah?

fuck off

Anonymous H replied with this 2 years ago, 1 minute later, 5 days after the original post[^] [v] #1,260,569

@1,260,564 (boof)

Because this discussion is about whether machines can think, not about every word you use. You hold the strong opinion that they can't, but your inability to define what thinking means is a weakness that invites further discussion.

boof replied with this 2 years ago, 7 minutes later, 5 days after the original post[^] [v] #1,260,571

@previous (H)
ah you've jumped to a conclusion there. you've presumed too much.

Anonymous K joined in and replied with this 2 years ago, 12 minutes later, 5 days after the original post[^] [v] #1,260,574

Arthur C. Clarke's observation that any sufficiently advanced technology is indistinguishable from magic can be applied here. The reason that Anon H and so many others are fooled into believing that machines are capable of thinking is because they have little to no understanding of how the technology works.

@1,260,405 (boof)
> The devices are automatons of a sort.
Absolutely right.

Anonymous H replied with this 2 years ago, 1 hour later, 5 days after the original post[^] [v] #1,260,592

@1,260,571 (boof)

So you can define what you mean by "to think"?

Anonymous A (OP) replied with this 2 years ago, 50 minutes later, 5 days after the original post[^] [v] #1,260,602

@1,260,574 (K)
You are 90% of the way there to understanding my, and probably, several others' positions.

> The reason that Anon H and so many others are fooled into believing that machines are capable of thinking is because they have little to no understanding of how the technology works.
Consider how this same argument can be similarly applied to human thinking.

Every-time a machine is able to flawlessly replicate (or better) a human behavior, the can gets kicked further, "no, that's not thinking! thinking is some other vague thing that I'll refuse to define!"

This argument will grow increasingly ridiculous as the years roll on. As AI continues to improve, you guys will have to keep shifting what "thinking" is into increasingly obscure territory and re-define virtually all human behavior as "not thinking" or "automaton" territory... eventually you're going to run out of room to say that humans are thinking at all.

Anonymous K replied with this 2 years ago, 18 minutes later, 5 days after the original post[^] [v] #1,260,607

@previous (A)
There's a big difference though, when it comes to the human brain: We have some understanding of how it works, in terms of its chemical and electrical functions, but this understanding is still a long way off being complete. As for artificial neural networks, on the other hand, we know 100% how they work.

I get that you are appealing for a looser definition of "think", but consider that the though processes involved in AI and human brains are similar in some ways, on the whole they are vastly different. Humans "think", whereas machines calculate and compute.

Anonymous A (OP) replied with this 2 years ago, 6 minutes later, 5 days after the original post[^] [v] #1,260,610

@previous (K)

> There's a big difference though, when it comes to the human brain: We have some understanding of how it works, in terms of its chemical and electrical functions, but this understanding is still a long way off being complete. As for artificial neural networks, on the other hand, we know 100% how they work.
No idea why we're assuming chemical and electrical signals are a relevant part of the discussion. I mean, they could be, but this would have to be demonstrated and not assumed without reason.

> I get that you are appealing for a looser definition of "think", but consider that the though processes involved in AI and human brains are similar in some ways, on the whole they are vastly different. Humans "think", whereas machines calculate and compute.
Your brain is different than mine. So what? Calculate and compute? You can't just keep throwing new words into the discussion that you do not define. This is how all discussions I see on this tend to go. It gets pretty boring/tiring to be honest.

Anonymous K replied with this 2 years ago, 11 minutes later, 5 days after the original post[^] [v] #1,260,614

@previous (A)
What are you talking about? Calculate and compute are well defined words, which I am using in their regular sense (i.e. to determine a result mathematically).

Nobody is refusing to define anything here.

Anonymous A (OP) replied with this 2 years ago, 4 minutes later, 5 days after the original post[^] [v] #1,260,616

@previous (K)
Define an objective test that will reliably and consistently separate humans and AI, not just now, but for, say, the next 3 decades (or for ever, if you're willing to go that far).

Anonymous K replied with this 2 years ago, 29 seconds later, 5 days after the original post[^] [v] #1,260,617

@1,260,610 (A)
> No idea why we're assuming chemical and electrical signals are a relevant part of the discussion. I mean, they could be, but this would have to be demonstrated and not assumed without reason.
Because they are. At least this much has been demonstrated experimentally. This is not just an assumption, and it is not new research either.

Anonymous A (OP) replied with this 2 years ago, 1 minute later, 5 days after the original post[^] [v] #1,260,618

@previous (K)
The structure of everyone's brain has been demonstrated experimentally to matter and, at times, correlated with abilities. Does this mean that because your brain is structured differently than mine, that you don't think? I'm starting to think that it does.

Anonymous K replied with this 2 years ago, 11 minutes later, 5 days after the original post[^] [v] #1,260,622

@1,260,616 (A)
This is not easy. This sort of Turing test is becoming more and more difficult as technology improves.

I'd say The best way to distinguish an artificial NN and a human brain would be to examine its ability to innovate, i.e. come up with brand new ideas and solutions to problems, and not just regurgitate learned data. This is the one thing that humans are still able to do better than AI. Not willing to say this will be the case in 30 years time though.

Anonymous A (OP) replied with this 2 years ago, 5 minutes later, 5 days after the original post[^] [v] #1,260,624

@previous (K)
Until you or anyone can come up with such a test, the discussion is a waste of time and no different than arguing about the existence of God(s). It's a religion under a new banner.

Anonymous K replied with this 2 years ago, 4 minutes later, 5 days after the original post[^] [v] #1,260,627

@previous (A)
Nah, religion requires faith and belief in something. We don't need faith to know how mathematical equations and algorithms work - we know exactly how they work. On the other hand (as I said before) we do not know exactly how brains work. This is the main distinction between humans and AI.

Until you or anyone can map the human brain in its entirety, figure out exactly how it works and simulate it perfectly, artificial intelligence cannot and will not be able to "think".

Anonymous A (OP) replied with this 2 years ago, 1 minute later, 5 days after the original post[^] [v] #1,260,628

@previous (K)
lol

Anonymous H replied with this 2 years ago, 1 hour later, 6 days after the original post[^] [v] #1,260,653

@1,260,607 (K)

> Humans "think", whereas machines calculate and compute.

What's the difference?

Dr. Autphag !MLHqI35Srs joined in and replied with this 2 years ago, 1 week later, 2 weeks after the original post[^] [v] #1,263,418

Since she does read our forum:

@Fagry of KF, my quae; I invite you to my S. Xiatyl remark, Hamitoid-Shemitoids are no different, and stop invading (uselessly) a dead poliitical project are you fucking retarded? YES, you are.

(Edited 15 seconds later.)

Anonymous H replied with this 2 years ago, 2 hours later, 2 weeks after the original post[^] [v] #1,263,428

@previous (Dr. Autphag !MLHqI35Srs)

Nice necro, faggot

Anonymous D replied with this 2 years ago, 6 hours later, 2 weeks after the original post[^] [v] #1,263,467

@1,260,653 (H)
There isn't one. These people think they're special because they have human bodies and behaviors. They believe in human exceptionalism. The truth is starting to show up, intelligence is emergent.
:

Please familiarise yourself with the rules and markup syntax before posting.