My man, you’re speaking sci fi, not what we currently have.

The biggest of current LLM models contains ~ the same number of parameters as we have neurons. It’s not a 1:1 mapping because parameters are closer to neuronal connections, but from a pure numbers standpoint we are operating at the scale where we can start creating true simulated intelligences, even if not human scale just yet.

This doesn’t mean current LLMs are that intelligent, just that it’s not sci-fi to think we could create a simulated intelligence now.

Furthermore, both philosophically and materially, the notion that consciousness cannot be computed is more than gaining traction.

Is it? Do you have any sources / do they have any explanation for why neurons can’t be simulated?

If humans ever make something with free will and volition, something that isn’t just doing things on command but has its own wants, sure. But we might never get there, and that’s a real possibility. Intelligence isn’t in solving equations but in imagining the math problems.

I mean, we’re talking about whether or not an AI could make music. If it creates a new song, with lyrics and music / a melody that never existed before, and people listen to it and sing it and dance to it and enjoy it, how would it not be music?

AI is not “making anything”, it’s regurgitating combinations of previous stuff on-command.

Even current day LLMs are doing more than just regurgitation, even if they fall far short of human intelligence.

And at a fundamental level, there’s no reason to think that simulated neurons running on computer chips can’t be as intelligent as us, if we can figure out the right way of wiring them so to speak.

There’s no inherent law of the universe that says that only biological humans can be intelligent and can thus create music.

That’s not a redefinition, lol, music is a human construct. Nature has lovely noises and birds chirp, and by itself, even if it constitutes notes and waves, it isn’t music.

A gorilla or ape can’t sing or make music? Could a neanderthal? Homo florientis? Homo erectus? What is it specifically about homo sapiens that give us the unique ability to make music and sing, that no other animal has?

Again, if you predefine music as being made by humans then you’re not engaging in a discussion or logical debate, you’re just arbitrarily setting goal posts to guarantee that you’re right.

People need to get over the idea that algorithms can’t be intelligent because they’re algorithms. Algorithms can model the behaviour of the neurons in your brain, meaning that they can model your brain and intelligence. We are obviously not there yet with LLMs, but just saying ‘numbers and math = not intelligent’ is quite frankly dumb and just shows that you don’t understand math, physics, biology, neuroscience, etc.

There is no human intentionality in AI-created music.

The entirety of your argument boils down to you arbitrarily deciding that music needs to derive from human intentionality.

That’s not an actual argument about whether or not AI is capable of creating music, that’s you redefining music to make sure the answer is no.

“Drawing the line” is a Thought Terminating Phrase, which is a concept worth reading and thinking about.

In the abstract / in your example, if you’re just “drawing the line” then no, that’s not a legitimate argument. It is literally just you saying “nope, I arbitrarily say this is different then this”.

However, if you can back up why one thing is different from the other, then it is valid to distinguish between them. Sometimes it can be worth dividing a system into chunks and drawing arbitrary lines rather not drawing any, but you should still be able to logically back up why it’s better to chunk things than not.

But in your example, it sounds like you don’t actually have a logical argument, just one based on you arbitrarily deciding that music can only be made by a human.

It can work for some people long term if they truly accept the person as they are and don’t get frustrated by them.

But often there will come a point, after the honeymoon has worn off, and after you’ve had sex too many times to count, where you suddenly realize that you’re not actually ever having conversations, you’re explaining stuff you found interesting to them, and getting nothing back. It can also get more frustrating depending on personality. If they’re always flying off the handle about stuff on social media and you have to explain why it’s not true, it will wear on you. If you always have to be the one to do everything, either because she doesn’t know how, or because you don’t trust her to do it right, then it will wear on you.

Personally that’s why I stress the idea of a partnership so much now. I want an equal partner who does everything I can do (or at least, most of it).

That being said, dating a 10/10 for a while and having a lot of incredible sex is also not an opportunity to necessarily turn down, especially depending on your age. If you’re young, go for it, have some fun, learn some lessons, have a lot of amazing sex, but get out early. If you’re older, then consider whether or not you can really be around that person, day in, day out, forever. If not bail now.

https://www.reddit.com/r/oddlysatisfying/comments/1q14kyj/playing_with_a_retro_floppy_disk_box/

They’re talking about the tactility of the format, not the actual data limits on it.

You could build SSDs today with the exact same tactility of floppy disks but with terabytes of storage.

deleted by creator

While the tester is blind as to which is which, the experimenter knows the construction of the machine and can presumably tell if it’s artificially constraining itself.

In the case of intelligences and neural networks that is not so straight forward. The humans and machines that are behind the curtain have to be motivated to try and replicate a human, or the test would fail, whether that’s because a human control is unhelpful or because the machine isn’t bothering trying to replicate a human.

I see what you’re saying but I think the problem is that you would need to test an AI while it’s unaware of being tested, or use a novel trick that it’s unaware of, to try and catch it producing non-human output.

If it’s aware that it’s being tested, then presumably it will try to pass the test and try to limit itself to human cognition to do so.

i.e. It’s possible that an AI’s intelligence includes enough human-like intelligence to completely mimic a human and pass a Turing test, but not enough to know to keep to those boundaries; but it’s also possible that it both knows enough to mimic us and enough to keep to our bounds, in which case the test then needs to be done in secret.

It is incorrect. No one is so good at math that they can design and build a full useful system on their own. That means they have to work with other people as part of a team. That means that their lack of social skills will doom or limit them.

It is also a mistake to equate psychopaths with having no social skills. They don’t have empathy, but they often have excellent social skills.

Are you in academia? This has not remotely been my experience in private industry.

Social skills pretty much trump all in engineering. If you write the most hyper efficient machine code, it will be a bitch to maintain and costs the team 10x as much going forward, but if you write code empathetically so that a normal human can pick it up, understand it, and fix it easily, then everyone will love you.

Same thing back when I was in physical engineering, you could have a brilliant idea for a design, but if you can’t communicate why it’s brilliant and why it’s worth getting everyone else to change to accomodate it, then it will get shot down.

Maybe reconsider throwing around words like “naiive” when your source is a Europol briefing document covering various threats at a high level with no stats or numbers.

Especially since if you actually dig into it, you’d find that Europe’s illegal gun trade comes partially from old military and police weapons from the Balkans / collapse of the Soviet Union, partially from the theft of legal firearms, partially from weapons that are imported (legally and illegally) from the US and Turkey, and minorly from weapons smuggled in from other war zones / 3d printed guns.

i.e. three out of four of the biggest sources of illegal guns in Europe are caused by legal firearm ownership, and one is the collapse of the Soviet Union.

The fact is that gun control works. Dislike that all you want but it doesn’t change the stats or reality of the world. Here in Canada the vast majority of gun crime is perpetrated using guns illegally smuggled from the US and another ~15% is from legal Canadian guns that were stolen. That’s not an argument that makes wide spread gun ownership look like a good idea.

Maybe so, but we live in a world where guns exist.

No, you live in a country that chooses to manufacture guns in response to people buying them, and you choose to actively perpetuate that by going and spending money buying guns and gun infrastructure, directly funding gun companies / their lobbies, and then by going online to try and spread that justification so that you can feel slightly less guilty about choices you’ve made that you know are wrong.

And where did they smuggle those guns from? Countries where it is easy to purchase legal firearms?

Even with 3D printed guns, literally no illegal entity is manufacturing them at scale, because that is still very, very traceable and catchable.

No, I misread your comment and thought it was from someone who had been in a situation where a gun would have made it better, rather than one where it had.

Interesting to note though, that at least one of those situations was caused by someone having easy access to firearms.

And some of you may be upvoting any plausible argument for gun ownership, even in the face of overwhelming objective evidence that it makes societies vastly unsafe.

Here’s the thing about guns and victimhood, access to guns causes far more victims then access to guns prevents, and it always inherently will. In that environment, a predator intent on committing a crime will always have one, and a victim only ever might have one.

If you rely on mutually assured destruction arguments, then you have armed and killing each other over road rage because humans are dumb emotional children who think they’re more mature then they are.

In this thread: Americans who literally cannot comprehend that they are not the only country on earth.

How did those situations resolve without having a gun?