Barack Obama: “For elevator music, AI is going to work fine. Music like Bob Dylan or Stevie Wonder, that's different”

L4sBot@lemmy.worldmod to Technology@lemmy.world – 491 points – Barack Obama: “For elevator music, AI is going to work fine. Music like Bob Dylan or Stevie Wonder, that's different”
musictech.com

Barack Obama: “For elevator music, AI is going to work fine. Music like Bob Dylan or Stevie Wonder, that's different”::Barack Obama has weighed in on AI’s impact on music creation in a new interview, saying, “For elevator music, AI is going to work fine”.

243

why do i care what obama feels about either of these

I very rarely care for what most 62 year olds have to say about the capabilities about the theoretical limits of computation.

This isn't much different.

If the 62 year old had studied computer science and had specialized in AI, I would listen closely to them.

But I definitely not care about a politician that has no idea about technology.

Unfortunately when it comes to medical experts, many ignore them and listen to their aunt with the healing crystals, or their buddy that skipped most of his high school science classes to go smoke behind the school instead...

I mean — he’s defending human creativity and he’s kind of right. AI can recreate variations of the things it is trained on, but it doesn’t create new paradigms.

People always says AI do create only variations but many successful TV shows are variations. I started watching sitcoms from the 70s and many things were copied/adapted in recent shows.

99% of everything people create is a variation.

Truly innovative anything is RARE.

There's just stuff and things people haven't thought to combine with stuff yet.

Yeah, also I think there is something about the human connection and communicating personal ideas and feelings that just isn't there with AI generated art. I could see a case for an argument that a lot of music today is recorded by artists who didn't write that music, and that they are expressing their own feelings through their performance of someone else's creation. And is it really all that different if an AI wrote something that resonated with an artist who ultimately performed it? Which for a good chunk of pop-culture regurgitations may be completely valid. But in my opinion, the best art, communicates emotion, which an experience unique to biology, AI might be able to approximate it, and sure there's a human prompting the AI who might genuinely have those feelings, but there's a hollowness to it that I struggle to ignore. But maybe I'm just getting older and will be yelling at clouds before long.

2 more...

If he got super wild and crazy by wearing a tan suit again to work would you?

2 more...

...so I've been on a shit load of elevators, and I don't recall a single one of them having music. For as common a trope as it is, you'd think elevator music would be more common in actual elevators.

It's like porn, they all used to have music, and now people still make jokes about how bad it was but it's just gone now

It's not as common as it used to be, but I think the point was kind of that you're not supposed to notice it?

Look into "muzak" (the style of music. Apparently it's also a brand according to Google), and some of Brian Eno's ambient albums like "Music for Airports" (which is definitely a bit more sparse than elevator music, which was often like smooth jazz versions of classic songs), but along similar lines.

I don't like to think I'm that old, and I 100% remember elevator music.

Edit: was possibly thinking of "musique concrete" rather than muzak.

I love Eno’s ambient music but it’s really distinct from the cheesy musak you’re referring to.

I may have been thinking of "musique concrete" rather than muzak

I've been in a few that played music (or more recently, ads) in them. But, yeah, it's like quicksand in that I was led to believe it would be pretty uniquitous.

or more recently, ads

Try this one trick to get me to take the stairs every single time.

I don't think I can actually recall one either.

Maybe in a department store or mall in the 80s. It was just so deliberately bland I never noticed when it became less common.

Some hotel elevators have it.

But yeah, I don't recall the last time I heard music in a residential elevator.

I worked in an office that installed music in the bathrooms. It wasn’t there for a long time, and then they added it. An email went out at one point instructing people to stop turning off the music (someone figured out where the Sonos controls were I guess). Someone at the top had decided it was IMPERATIVE to have something to listen to other than the coworker grunting next to you.

There is no way this ages well.

I think the statement was more about the impact, which will depend on each person's subjective experience

Personally I agree. Even if AI could produce identical work, the impact would be lessened. Art is more meaningful when you know it took time and was an expression/interpretation by another human (rather than a pattern prediction algorithm Frankenstein-ing existing work together). Combine that with the volume of AI content that's produced, and the impact of any particular song/art piece is even more limited.

I'd say art is more meaningful when it's a unique experience. It's like those myths about glassmakers being killed blinded after the cathedral is finnished so that no one can replicate the glass color... without the killing.

People are social, if enough people feel the same way about one thing it'll succeed. It doesn't matter where it came from or how it was made, like how people can still admire and appreciate nature. Or maybe the impact will be that it reduces all impacts. Every group and subgroup might be able to have their own thing.

I don’t know. I think Obama kind of nailed it. AI can create boring and mediocre elaborations just fine. But for the truly special and original? It could never.

For the new and special, humans will always be required. End of line.

At this point I want a calendar of at what date people say "AI could never" - like "AI could never explain why a joke it's never seen before is funny" (such as March 2019) - and at what date it happens (in that case April 2022).

(That "explaining the joke" bit is actually what prompted Hinton to quit and switch to worrying about AGI sooner than expected.)

I'd be wary of betting against neural networks, especially if you only have a casual understanding of them.

I mean the limitations of LLMs are very well documented, they aren't going to advance a whole lot more without huge leaps in computing technology. There are limits on how much context they can store for example, so you aren't going to have AIs writing long epic stories without human intervention. And they're fundamentally incapable of originality.

General AI is another thing altogether that we're still very far away from.

Nearly everything you wrote is incorrect.

As an example, rolling context windows paired with RAG would easily allow for building an implementation of LLMs capable of writing long stories.

And I'm not sure where you got the idea that they were fundamentally incapable of originality. This part in particular tells me you really don't know how the tech is working.

A rolling context window isn't a real solution and will not produce works that even come close to matching the quality of human writers. That's like having a writer who can only remember the last 100 pages they wrote.

The tech is trained on human created data. Are you suggesting LLMs are capable of creativity and imagination? Lmao - and you try to act like I'm the one who's full of shit.

That's like having a writer who can only remember the last 100 pages they wrote.

That's why you pair it with RAG.

The tech is trained on human created data. Are you suggesting LLMs are capable of creativity and imagination?

They are trained by iterating through network configurations until there's diminishing returns on how accurately they can complete that human created data.

But they don't just memorize the data. They develop the capabilities to extend it.

So yes, they absolutely are capable of generating original content that's not in the training set. As has been demonstrated over and over. From explaining jokes not found in the training data, solving riddles not found in it, or combining different concepts to result in a new synthesis not found in the original data.

What do you think it's doing? Copy/pasting or something?

I think, it will eventually become obsolete, because we keep changing what 'AI' means, but current AI largely just regurgitates patterns, it doesn't yet have a way of 'listening' to a song and actually judging whether it's good or bad.

So, it may expertly regurgitate the pattern that makes up a good song, but humans spend a lot of time listening to perfect every little aspect before something becomes an excellent song, and I feel like that will be lost on the pattern regurgitating machine, if it's forced to deviate from what a human composed.

I have seen a couple successful artists in different genres admit to using AI to help them write some of their most popular songs, and describe it's use in the songwriting process. You hit the nail on the head with AI not being able to tell if something is good or bad. It takes a human ear for that.

AI is good at coming up with random melodies, chord progressions, and motifs, but it is not nearly as good at composing and producing as humans are, yet. AI is just going to be another instrument for musicians to use, in its current form.

Yeah, I do imagine, it won't be just AIs either. And then, it will obviously be possible to take it to an excellent song, given enough human hours invested.

I do wonder, how useful it will actually be for that, though. Often times, it really fucks you up to try to go from good to excellent and it can be freeing to start fresh instead. In particular, 'excellent' does require creative ideas, which are easier for humans to generate with a fresh start.
But AI may allow us to start over fresh more readily, if it can just give us a full song when needed. Maybe it will even be possible to give it some of those creative snippets and ask it to flesh it all out. We'll have to see...

As someone who is doing software engineering and my company jumped on AI bandwagon and got us GitHub Copilot. After using it for a while I think overall experience is actually net negative. Yes, sometimes it gets things right, sometimes it provides a correct solution, but often I can write much more concise code. Many times it provides code that looks like it is correct, but after looking in more detail it actually is wrong. So now I'm need to be in guard what code it inserts, which kills all the time that it supposedly saved me. It makes things harder because the code does look like it might work.

It is like pair programming with a complete moron that is very good at picking patterns and trying to use them in following code. So if you do a lot of copy and paste I think it will help.

I think this technology can make bad programmers suck less at programming. I think the LLM problem is that it was trained with existing works and the way it works is that its goal is to convince other human that the result was created by another one, but it isn't capable to do any actual reasoning.

Wow, my experience has been pretty much the exact opposite of this. Copilot is amazing and I'd rather not go without it ever again

Edit: for the life of me I'll never understand people. This comment got a bunch of downvotes and yet some douchebag who blindly accuses me of being bad at my job gets upvoted. Fuck people.

What language you program in and what kind of code you develop? Before Copilot were you frequently searching answers on stackoverflow?

Typescript, JavaScript, php, bash, scss/css... And isn't every dev on SO or at least a search engine with some frequency?

I don't actually think the reason I like it is dependent on the language at all. The reason I like it is that it will often basically notice what I'm doing and save me from typing a repetitive 3-5 line block. Things like that and if I can't remember a specific syntax, I've found that I can write a comment saying what the following code will do and boom, suddenly copilot writes a version of that code close to what I would've written.

I mean you're right that it can write stuff that doesn't work, I just find that I can usually filter that out pretty quickly. The times I can't, I'm a bit stuck anyway and it's worth a shot to try their mysterious solution. But since I always treat its solutions with skepticism I haven't been bitten yet.

For me, copilot just takes the monotony out of the job. Instead of spending as much time writing boring stuff I get to focus on the more interesting parts

Maybe you aren't that good at writing code

Maybe you aren't that good at being a human, this comment being good evidence of that

Ignore them. At some point you gotta realize most people are losers trying to bring others down with them.

Do what works for you :)

I appreciate this comment. You inspire me to not only ignore more assholes, but maybe I'll also be one myself less often :)

Ill blindly accuse you of being bad at your job too, bud.

Thanks for block request. Appreciate reducing douchebags in life

Do people actually care what Obama has to say about AI? I'm just having a hard time seeing where his skillset overlaps with this topic.

Probably as much as I care about most other people's thoughts on AI. As someone that works in AI, 99% of the people making noise about it know fuck all about it, and are probably just as qualified as Barack Obama to have an opinion on it.

What do you do exactly in AI? I'm a software engineer interested in getting involved.

I work for Amazon as a software engineer, and primarily work on a mixture of LLM's and compositional models. I work mostly with scientists and legal entities to ensure that we are able to reduce our footprint of invalid data (i.e. anything that includes deleted customer data, anything that is blocked online, things that are blocked in specific countries, etc). It's basically data prep for training and evaluation, alongside in-model validation for specific patterns that indicate a model contains data it shouldn't have (and then releasing a model that doesn't have that data within a tight ETA).

It can be interesting at times, but the genuinely interesting work seems to happen on the science side of things. They do some cool stuff, but have their own battles to fight.

That sounds cool, I've had roles that were heavy on data cleansing, although never on something so interesting. What languages / frameworks are used for transforming the data, I understand if you can't go into too much detail.

I did wonder how much software engineers contribute in the field, it's the scientists doing the really interesting stuff when it comes to AI? Not surprisingly I guess 😂

I'm a full stack engineer, I was thinking of getting into contracting, now I'm not so sure, I don't know enough about AI's potential coding capabilities to know whether I should be concerned about job security in the short, or long term.

Getting involved in AI in some capacity seems like a smart move though...

We do a lot of orchestration of closed environments, so that we can access critical data without worry of leaks. We use Spark and Scala for most of our applications, with step functions and custom EC2 instances to host our environments. This way, we build verticals that can scale with the amount of data we process.

If I'm perfectly honest, I don't know how smart a move it is, considering our org just went through layoffs. We're popular right now, but who knows how long for.

It can be interesting at times, but to be honest if I were really interested in it, I would go back and get my PhD so I could actually contribute. Sometimes, it feels like SWE's are support roles, and science managers only really care that we are unblocking scientists from their work. They rarely give a shit if we release anything cool.

Tbf everyone is entitled to have an opinion, including Obama

There is a tad bit of difference between caring about an opinion and tolerating one. Obama's opinions on AI are unqualified pop culture nonsense. They wouldn't be relevant in an actual discussion that would cite relevant technical, economical and philosophical aspects of AI as points.

Sure, care about it or don't, I don't care. It was the "being qualified to have an opinion" bit I didn't like. I don't have to qualify to have an opinion and I can write an opinion piece and sure enough, less people will read it than Obama's. I might not be qualified to teach on that subject but everyone is qualified to build one's own opinion.

But maybe that's just overly pedantic on my side. You are qualified to have a different opinion.

I know this was once said about the automobile, but I am confident in the knowledge that AI is just a passing fad

Why? It's a tool like any other, and we're unlikely to stop using it.

Right now there's a lot of hype because some tech that made a marked impact of consumers was developed, and that's likely to ease off a bit, but the actual AI and machine learning technology has been a thing for years before that hype, and will continue after the hype.

Much like voice driven digital assistants, it's unlikely to redefine how we interact with technology, but every other way I set a short timer has been obsoleted at this point, and I'm betting that auto complete having insight into what your writing will just be the norm going forward.

It's just a Chinese room dude, it doesn't actually do anything useful

The Chinese room argument doesn't have anything to do with usefulness. Its about whether or not a computer that passes the turing test is conscious. Besides, the argument is a ridiculous one to begin with. It assumes that if a subcomponent of a system (ie the human) lacks "understanding", then the system itself (the human + the room + the program) lacks understanding.

Anything else aside, I wouldn't be so critical of the thought experiment. It's from 1980 and was intended as an argument against the thought that symbolic manipulation is all that's required for a computer to have understanding of language.
It being a thought experiment that examines where understanding originates in a system that's been given serious reply and discussion for 43 years makes me feel like it's not ridiculous.

https://plato.stanford.edu/entries/chinese-room/#LargPhilIssu

What?

At best you're arguing that because it's not conscious it's not useful, which.... No.
My car isn't conscious and it's perfectly useful.

A system that can analyze patterns and either identify instances of the pattern or extrapolate on the pattern is extremely useful. It's the "hard but boring" part of a lot of human endeavors.

We're gonna see it wane as a key marketing point at some point, but it's been in use for years and it's gonna keep being in use for a while.

A system that can analyze patterns and either identify instances of the pattern or extrapolate on the pattern is extremely useful. It’s the “hard but boring” part of a lot of human endeavors.

I agree with most of what you're saying here, but just wanted to add that another really hard part of a lot of human endeavors is actual prediction, which none of these things (despite their names) actually do.

These technologies are fine for figuring out that you often buy avocados when you buy tortillas, but they were utter shit at predicting anything about, for instance, pandemic supply chains....and I think that's at least partially because they expect (given the input data and the techniques that drive them) the future to be very similar to the past. Which holds ok, until it very much doesn't anymore.

I’m sorry, they aren’t good at predicting?

My man, do you have any idea how modern meteorology works?

A ton of data gets dumped into a ton of different systems. That data gets analyzed against a bunch of different models to predict forecasts The median of al those models is essentially what makes it into the forecast on the news.

Well, I would disagree that they don't predict things. That's entirely what LLMs and such are.

Making predictions about global supply chains isn't the "hard but boring" type of problem I was talking about.
Circling a defect, putting log messages under the right label, or things like that is what it's suited for.

Nothing is good at predicting global supply chain issues. It's unreasonable to expect AI to be good at it when I is also shit at it.

They make probabilistic predictions. Which are ok if you're doing simple forecasting or bucketing based upon historical data, and correlates and all of that.

What they are crappier about is things that are somewhat intuitively obvious but can't be forecasted on the basis of historical trends. So, like new and emerging trends or things like panic buying behavior making it so the whole world is somehow out of TP for a time.

I'd argue that relying solely on "predictive analytics" and just in time supply chains aggravated a lot of issues during the big COVID crunches, and also makes your supply chain more brittle in general.

All predictions are probabilistic.

AI indeed isn't great at modeling complex or difficult to quantify phenomenon, but neither are people.

Our recent logistical issues are much more based on the frailty of just in time supplying than the methods we use to gauge demand. Most of those methods aren't what would typically be called AI, since the system isn't learning so much as it's drawing a line on a graph.

We didn't actually run out of toilet paper, people just thought we did and so would buy all of it if they saw it in the shelves. It's a relatively local good, so it didn't usually get caught up in the issues with shipping getting bogged down, it's just that people chose to override the model that said that stores should buy five trucks full of TP because it would fill their warehouse and they were worried they'd be stuck with the backlog.

All predictions are probabilistic.

Eh, not really. All math / model based predictions are probabilistic. There's other ways to make predictions, and not all of them are based on math, and they might be wrong more often than a probabilistic model, but they exist.

AI indeed isn’t great at modeling complex or difficult to quantify phenomenon, but neither are people.

Sure, fair enough, but there are times where a computer model is missing obvious context, and it's those times that I think we have to pay attention to.

The current industry adoption patterns seem to veering pretty close to "the computer did that auto-layoff thing" from Idiocracy in my opinion.

You not having a job where you work at a level to see how useful AI is just means you don't have a terribly important job.

What an brain drained asshole take to have. But I've seen your name before in my replies and it makes sense that you'd have it.

AI is useful for filling out quarterly goal statements at my job, and boy are those terribly important... 😆

Absolutely not. We need to learn the difference between intelligence and expertise. Is Obama an intelligent person? Of course. Is he allowed to have and voice an opinion? Sure, it's a free country. Does that mean that his opinion is informed by expertise and should dictate peoples actions and therefore the direction of an industry? No.

This is the same logic that allows right wing ideologues to become legitimate sources of information. A causal interest in a topic is NOT the same as being an industry expert, and the opinions of industry experts should be weighted far heavier in our minds than people who "sound like they know what they're talking about".

This is the same logic that allows right wing ideologues to become legitimate sources of information. A causal interest in a topic is NOT the same as being an industry expert, and the opinions of industry experts should be weighted far heavier in our minds than people who “sound like they know what they’re talking about”.

And your logic is the same followed by government agencies when they effectively agree to regulatory capture because all of the industry experts work at this company, so why not just let the company write the rulebook? 🤔

I personally don't believe we need "industry experts" in every new, emerging type of tech to be the sole voices considered about them because that's how we largely arrived at the great enshitterment we're already experiencing.

Edit: It's really quite a baffling take (given a moment's thought) that the big problem and/or a large problem facing America is that we aren't cozy enough with "industry experts". Industry practically write the policy in this country, and the only places where we have any kind of great debate (e.g. net neutrality, encryption) is where there are conflicting industry concerns.

For a second or two I was like: " why the f*** is Obama talking about Weird Al making elevator music?"

I'm just a dude who does general labor and have lots of insights about AI just because I'm interested and smart. People tend to come to me just to hear what I have to say.

Now look at Obama. He's all of that and much more in the eyes of a society that's put Obama in the spotlight. He can talk about totally boring stuff and people will still respect his opinion.

But do we really need AI to generate art?

Why can't AI be used to automate useful work nobody wants to do, instead of being a way for capital to automate skilled labor out of high-paying jobs?

Because AI is unpredictable. Which is not a big issue for art, because you can immediately see any flaws and if you can't, it doesn't matter.

But for actually useful work, you don't want to find out that the AI programmer completely made up a few lines of code that are only causing problems when the airplane is flying with a 32° bank angle on a saturday with a prime number for a date.

We currently have the same problem with human programmers. That's why good companies always test the shit out of their code.

It's virtually guaranteed that at some point, robots and/or AI will be capable of doing almost every human job. And then there will be a time when they can do every job better than any human.

I wonder how people will react. Will they become lazy? Depressed? Just have sex all the time? Just have sex with robots all the time?

It depends if the government introduces universal basic income or not. If they do I couldn't care less if I don't have a job. Any reason I have a job is so that I have money. I don't do it so I have some kind of fulfillment in my life because it isn't a fulfilling job.

Just have sex all the time?

I'm confused about how this one tracks. Is the AI going to make me more attractive or is it just going to lower everyone else's standards?

Like all other animals, humans evolved to be more likely to procreate. There is an argument that all of that other stuff we do is just in support of procreation. But in a way, it's also a distraction and an impediment to procreation. It just so happens that we've been unable to avoid doing that other stuff so far.

Even animals do other things though they play and socialise.

Personally if AI makes working irrelevant I'm just going to spend most of my time in Disneyland.

It'll be funny if we avoid post-scarcity because we want people to keep working for each other.

You talk like AI I'd a singular entity that can only do one thing?

Why should we stifle technological progress so people can still do jobs that can be done with a machine?

If they still want to create art, nobody is stopping them. If they want to get paid, then they need to do something useful for society.

Nobody's calling to stifle technology or progress here. We could develop AI to do anything. The question is what should that be?

There's a distinction to be drawn between 'things that are profitable to do and thus there isn't any shortage of' and 'things that aren't profitable and so there's a shortage of it' here. Today, the de facto measure of 'is it useful for society?' seems to be the former, and that doesn't mean what's useful for society, it's what's usefuI for people that have money to burn.

Fundamentally, there isn't a shortage of art, or copy writers, or software developers, or the things they do- what there is, that AI promises to change, is the inconvenient need to deal with (and pay) artisans or laborers to do it. If the alternative is for AI vendors to be paid instead of working people, is it really the public interest we're talking about, or the interests of corporate management that would rather pocket the difference in cost between paying labor vs. AI?

AI is an enabler. I have not patience for sitting and drawing for hours on end to make extremely detailed art but I'm a creative individual and would love to have the power to bring my ideas into reality. That's what AI art does.

The problem with that, of course is it means that if I'm really serious about an idea I won't be paying some artist(s) to make it happen. I'll just whip open an AI art prompt (e.g.Stable Diffusion or any online AI art generators) and go to town.

It often takes a lot of iteration and messing with the prompt but eventually you'll get what you want (90% of the time). Right now your need a decent PC to run Stable Diffusion (got 8GB of VRAM? You too can generate all the AI images you want 👍) but eventually people's cell phones in their pockets will be even better at it.

Civitai is having a contest to make a new 404 error page graphic using AI. Go have a look at some of the entries:

https://civitai.com/collections/104601

I made one that's supposed to be like the Scroll of Truth meme:

Scroll of Truth meme 404 error page

I made that on my own PC with my limited art skills using nothing but automatic1111 stable diffusion web UI and Krita. It took me like an hour of trying out various prompts and models before I had all the images I wanted then just a few minutes in Krita to put them into a 4-panel comic format.

If I wanted to make something like that without AI it just would never have happened.

Not that it really matters in this case, but AI art just seems inconsistent in silly ways. That girls shirt changes each frame, her hair gets more braided, and the 3rd frame has 2 left hands. I guess at first glance you don't really notice, but it's not hard to spot and it hurts my brain once I do.

Also her face expression is so generic it doesn't convey the meaning like the original did.

Honestly if I didn't know the original I wouldn't understand the point of this one.

I almost immediately noticed the shirts and frame 4s 8 fingered hand

1 more...
1 more...

But do we really need AI to generate art?

No, but we want it to. It's probably only a matter of time untill AI can do better anything that humans can, including art. Now if there's an option to view great art done by humans or amazing art done by AI I'll go with the latter. It can already generate better photographs than I can capture with my camera but I couldn't care less. Takes zero joy out of my photography hobby. I'm not doing it for money.

We don't need it. It's just cool tech. I've messed around with stable diffusion a lot and it's a cool tool.

I don’t think it’s really helpful to group a bunch of different technologies under the banner of A.I. but most people aren’t knowledgeable enough to make the distinction between software that can analyze a medical scan to tell me if I have cancer and a fancy chat bot.

3 more...
4 more...

Why would people think he knows anything about AI?

Because we often look up to public figures for stuff they are not qualified to comment on.

If he's an unqualified bystander, then what the fuck are you?

I'm always surprised that the people with all the answers only share them with thirty other assholes on the Internet.

I'm confident a 14 year old can write their own AI, maybe even a smart 10 year old.

Here's instructions for kids:

https://youtu.be/XJ7HLz9VYz0?si=1QN3fqT03HSMufib

You think Obama can't wrap his head around a little algebra?

Why, when speaking intelligently and thoughtfully in the subject, is he so wrong in his assessment, when you, in one lazy sentence, are so right?

I'm really worried about would-be wise people just throwing in the towel cause they don't know how much better they could be with a little discipline, and settle for being clever here and there.

As far as I know, Obama has nothing to do with IT and doesn't have a big interest in it. A lot of people on here are probably more qualified than he is when it comes to these topics simply because they spent a lot of their free time learning about it.

1 more...
7 more...

Because he's a world leader and AI programs are answering search engine queries with what you want to hear now, not actual answers. Ain;t no way hes unaware that.

Because you can teach a teen to do it in two weeks. He was a constitutional law professor, as well as the first elected African-American president in the United States. I learned LLMs in a couple months and I never used a comp until 2021. Why are you gatekeeping?

Using the end product and having any idea how it works are two VERY different things.

I agree, my argument is that both aren't challenging for even the average person if they really want/need to understand how these models produce refined noise informed by human patterns.

There are electricians everywhere you know.

This isn't a random person thoughtlessly yelling one-sentence nonsense pablum on the Internet like you.

You think this person can't understand something as straightforward as programming, coming from law?

https://en.wikipedia.org/wiki/Barack_Obama

Please link your Wikipedia below 🫠

It's a bit more complicated than you're making it out to be lmfao, there's a reason it's only really been viable for the past few years.

The principles are really easy though. At its core, neural nets are just a bunch of big matrix multiplication operations. Training is still fundamentally gradient descent, which while it is a fairly new concept in the grand scheme of things, isn't super hard to understand.

The progress in recent years is primarily due to better hardware and optimizations at the low levels that don't directly have anything to do with machine learning.

We've also gotten a lot better at combining those fundamentals in creative ways to do stuff like GANs.

Why are you acting like it's at all difficult to understand?

AI researcher (PhD) here and for what it's worth, Obama got it extremely right. I saw this and went "holy shit, he gets it"

Yeah I dont believe you at all. I got my master in AI 8 years ago and have been working in the field ever since and no one with any knowledge would agree with you at all. In fact I showed a couple of my colleagues the headline of this article and they both just laughed.

If you don't think ai will get there and surpass everything humans have done in the past, you should change career.

I'm saying this because I do this for a living. It has become obvious to everyone in research (for example - https://arxiv.org/abs/2311.00059) that "AI"s don't understand what they are outputting. The secret sauce with all these large models is the data scale. That is, we have not had real algorithmic breakthroughs - it's just model scale and data scale. So we can make models that mimic human language and music etc but to go beyond, we need multiple fundamentally different breakthroughs. There is a ton of research attention now so it might happen, but it's not guaranteed - the improvements we've seen in the past few years will plateau as data plateaus (we are already there according to some, i.e we've used all the data on the Internet). Also, this - https://arxiv.org/abs/2305.17493v2

You do it for a living and you can't even understand what a general ai is. Alas I long since understood that mostly everyone is profoundly incompetent at their own jobs.

8 more...

While reassuring for many to hear, that's only going to be true for so long. Eventually it's going to be real fucking good at making "real" music. We need to be preparing for those advancements rather than acting like they'll never come.

I feel very reassured to hear that from the AI expert / musical virtuoso himself, 62 year old, former United States President Barack Obama.

Right? That said, I care about his opinion on AI about as much as I care about the opinions of all the anonymous loudmouths on Lemmy.

Yeah Barrak should just shut up. That idiot hasn't even listened to music before.

To make "real" music, AI will probably need a lot of help. Image generators and chat bots seem to have their own, very boring style. I've seen videos of artists using AI tools in their workflow, but it's still a very involved process. I think it will just be another tool for musicians and sound engineers.

I'm the immediate term, yeah I 100% agree. However, I'm not thinking we bank on that being true forever.

Obama must not have heard There I Ruined It.

While I agree, it's also the case that those ...Creations... are extremely human directed. As far as I know the maker is not only training the models for the voices, but also specifying each output word, and then its timing and pitch(s)

And of course placing the siren whistle.

One of my jobs involved updating blogs for small businesses. I had a Shutterstock subscription for the images that goes along with these blog posts. For this task, I think AI generated images work a lot better than stock photography.

There's some recruitment company advertising jobs on LinkedIn. All the pictures are clearly AI generated and they're terrifying. Uncanny Valley freaks grinning at you from your screen.

you can already api into chatgpt and dall-e 3 as one cohesive service, and make a system in an afternoon's work that reads the article, decides on a thumbnail, and automatically generates one. the whole thing costs like 8 cents per article.

You could also self host Stable Diffusion to save some money

Edit: Or is it free to use with ChatGPT Premium (or whatever it's called)? Then that would actually be cheaper

no, the premium stuff doesn't give you api access. which is total bs, but yeah, it's only for that grey interface. (i'm also quite salty that the playground has no easy to access image inputs but that's beside the point)

you're completely right about self-hosting sd, it's just a matter of prompting. sd workflows tend to get a little more experimental but i guess you could still make chatgpt write a few prompts that are close to correct and just manually rerun if an image failed

Maybe ChatGPT knows how to write SD prompts at this point. You could try just telling it to generate a prompt specifically for SD.

I still need to set up my computer to do the AI generated images.

My office is using AI to do this years xmas card. It's pretty cool.

Definitely need more people to tell me about ai and what it will be capable of. Make a daily show so that every shitty celebrity can tell us about ai, there might still be plenty of word combinations that haven't been used!

Elevator music as well as the mainstream music that majority of people listen to like pop etc.

That music is already very formulaic and almost as if it is generated by Ai.

Already likely to be untrue, but honestly I'd happily sign up for a world wear "hold music" isn't the same 20sec loop of shit jazz

That actually might make elevator and phone hold music survivable - continual compositions that never repeat

Cheaper to generate 30s and loop it

That's the current state of fuckery and makes self immolation a tempting option

2 more...

RIP Kenny G

He's survived many mockeries over the years haha

Okay, I love the elevator music idea as a gag in media.

But I've never been in an elevator that has ever played music, and I can completely understand why. Elevator music sounds obnoxious.

It tends to be in upscale hotels, generally around the holidays, but isn't incredibly common anymore.

If I think back on it, I'm not sure I have heard it since the 90s.

Wait a minute. Me neither. I never questioned it, yet....

Lots of people play lowfi hiphop as background noise. It's the modern equivalent of late 80s smooth jazz elevator music.

Where are they getting the training data from for AI music models? I guess it's the same issue as art and language models, but wouldn't they need to only use royalty free music?

Kidz Bop and Max Martin's drunk voicemails

In before obligatory republican outrage and 24x7 media coverage explaining how this comment will doom democrats in 2024

It's more or less only (that is mainly) useful for building components that you then use in your man-made tracks. It's a tool, just like AI image generators are tools albeit there the replacement use-case is substantial. AI-generated voice also needs to be considered in this context I think.

Yeah generative music has been a thing for a long time, Brian Eno is probably the household name recognizable for generative compositions, but most sequencers have had randomization elements built in for a long time now. I use one where you feed it a scale of notes and can define the chance a certain note will play and chances around the quality of the note like duration, velocity, etc. Even my entry level MicroFreak has a randomization option which you can use to get musical ideas from. There's some cool eurorack modules like Mutable Instruments Grids which function like this for drum sequencing, where you have this axis to explore and can control via an lfo if you want.

I realize generative and AI are a technically different, I think AI is much better at "can you create a synth preset to make x sound" or "write a specific genre of melody/chord progression/etc." It's a lot better at factoring in the broader context.

It's reassuring that this opinion is based on many years of experience reading scientific papers, implementing these models and following the trends closely!

I was really hoping the comments would be some sort of “what the fuck does he know?”

I scrolled a bit and there are many like these, yep

Wow, a word from a global expert on AI, Barrack Obama. I hope he’s a bit better at it than he is at world peace!

No, people need to start being creative and to make their own shit again.

No....you agree with him?

I mean no, it shouldn't be used to make any music, least of all shitty elevator music.

Who else should made elevator music then?

Actual human beings. Or just use music already made by people.

But why, I don't think any artist has ever thought:

"Creating elevator music is so fulfilling! I would never take a different job and express myself in unique and creative ways, bland repeating background noise music is my passion!"

Believe it or not, they do.

Like take hold music for example: the hold music you hear most of the time is actually called Opus No. 1 and was written by two teens in their garage in 1989, and when one of them went to work for Cisco ten years later, he offered it to them for their hotline. True facts, look it up.

Don't ask me how the hell I just know this shit. 🤦

I mean, people are doing that. It's just that it'd be a hell of a lot cheaper to program an AI to be creative than to pay actual creative people.

AI is new and already performing in office to a second year law student level.

That's what I got from this.

Also the Secret Invasion intro looked really cool and I hate it but it did. When it comes to depth and meaning it may be a while yet before AI can simulate or bullshit it's way through that but for now it can definitely create interesting and noteworthy pieces.

What is notable that there is a unique culture of making human covers of popular AI based songs. Especially in the vocaloid scene. Same with auto tune and similar tools.

AI isn't going to take the jobs of musicians. It'll just give them more material to work with because people will always seem out "authentic" versions of things they like.

"AI can't do [thing]" *AI does thing* "Well AI can't do [other thing]" *AI does other thing* "Yeah, well AI doesn't have soul (or some other nebulous thing)"

We can't just dismiss rapidly improving AI and ignore the coming consequences. Things like "soul" or any other element that is considered human never mattered when it came to commercial music. A lot of radio hits are quickly-produced trash, with perhaps a catchy tune, and little in the way of meaningful lyrics. If it's considered "good enough" (i.e. it makes money) then that's all that is needed for a company. Don't fool yourself into thinking the artistry ever mattered in this context.

This is the reality a lot of people don't want to face, a huge amount of popular music is just established patterns put together in the way music school teaches is the right way then pushed with endless hype from the money hungry corporations.

We are absolutely going to see a kid in his bedroom use AI to make a concept album that resonates with people and garners popular acclaim - I don't know when it will happen but I wouldn't be at all surprised if it's this decade.

People used to say that same nonsence about samplers and drum machines, it's the same trend as when people said Shakespeare isn't poetry because he hadn't been trained in Latin and Greek enough or Cezanne wasn't a real artist, or a million other times people have confidently declared we're at the end of history and nothing new will ever be good.

What people need to remember is when they see Rick Astley singing his heart out it's lyrics written by Pete Waterman and music from session artists put together by university trained professionals to shift units - soulless music has been a thing for a long time and no one ever cared before.

Obsessive kids have been making great music in their bedroom for decades now, aphex twin gets industry acclaime for twiddling some knobs and so will whoever gets popular doing the equivalent interesting things with AI.

He's obviously never heard "I'm still standing" by Adolf Hitler.

Those are just voice replacements

I hadn't before reading this comment either, which prompted me to look for it on YouTube.

Of course, Adolf Hitler couldn't actually speak English, and the English generated by that doesn't sound very much like a native speaker of Austrian German would normally speak English. I am a native speaker of Austrian German and if I sing "I'm Still Standing" it doesn't sound like that at all.

I mean Obama is not wrong, but I hope the rest of his thoughts are that it will be as good as the artists he quoted as it absolutely will one day. Unsure when as there is an uncanny valley to cross. He did say this right and it was not just a get off my front lawn comment…right?

He's wrong in the sense that Bob Dylan and Stevie Wonder are just as much cultural reflections as they are artists. Popular art and celebrity are effectively inseparable. This is why each generation often doesn't "get" the popular art of others.

The reality is that there's nothing stopping an AI from becoming a celebrity, and I'm a bit surprised it hasn't happened yet. Once that happens, the art it produces will be lauded, and people who are not the direct participants in mainstream popular culture will be confused and in some cases upset or offended by it. And the wheel will continue to turn.

Very good points. Art is in the eye/ear of the beholder sort of thing and our taste is for sure influenced by our culture/generation.

LMAO, what an absolutely god awful take.

Ill see you in 2 years when we have our first AI Grammy winner

This thread is as much evidence of our failed educational system as is the entire state of Mississippi

Wow, coming outta nowhere to sideswipe Mississippi for no apparent reason, great work you’re doing here.

Old man doesn't understand how technology progresses. Tonight at 10pm.

It's hilarious to me that young people think "AI" is intelligent or good at what it does. It's a fucking toy. I've made my own little GANs and Markov chains and the are thousands of empty little parlor tricks we've invented to separate the young from their money.

I saw a thread where a few zoomers were going wild over an AI loli generated porn image that looked like an animal balloon, all oiled and shiny, vagina sideways, face of an eight year old smiling for a school photo, huge veiny cocks blasting gallons of cum onto a tiny kawaii face:

"Omg so fucking hot 🥵 I'm fucking diamonds!"

Your dumb ass will be listening to a "song" simulacrum constructed out of 100,000 product jingles and Max Martin B-sides, jacking off to deranging porn, having full blown emotional conversations with bots and crying when that bot's creator gets busted for CP.

Anyone listening to false music that's overly Fourier transformed, with impossible instrument voices and chords that make it unlistenable to actual musicians and then trying to make people feel ashamed for having a soul is just waving a massive red flag for everyone to see that says "I have no idea what it means to be a human with skin in the game".

That movie Demolition Man totally called it in the 90s btw that not only would you all be listening to commercial jingles all the time but defending them and singing their "lyrics" loudly.

Thank you for loudly identifying yourself as yet another Elon-like who thinks they know so much more than Barack fucking Obama. You and yourself should get a room and listen to that shit all day and see who you become. If you can't create your own, you kids will get the "art" you deserve and it will cost A LOT 😂

Following for the laughs!

This post is disturbing and weird, you have a very odd and creepy view of the world.

Obama is just a random person voicing a safe and middle of the road politically neutral option in response to a boring question, I wouldn't put too much stock in his kneejerk responce to a new technology.

Anyway all I wanted to ask is what do you mean by it costing A LOT? The price of creating media has fallen insanely in the last five decades and is only continuing to plumet - processor power likewise, it's very likely we'll be able to not just run the models on consumer graphic cards as we can now but run them as background processes on our phones without noticing slowdown. Just like all other modern content we're going to see the free stuff that people make and share displace the hacky old corporate stuff - people that like history documentaries don't watch history channel they which YouTube and a couple of ken Burns level creations, same with science and tech and so many other types of content.

When a kid in their bedroom can make a movie that looks as good as marvel and has a powerful soundtrack that carries the action and moves the heart no one is going to care if they created the music in a weekend talking to AI or hired a generic studio musician to string together some standard progressions and pachelbell melodies - likely they'll prefer the ai output anyway because it'll fit the artist vision more than a hired chord ever could.

There are going to be creative geniuses that use AI to make amazing things and at some point you're just going to have to accept that.

2 more...

I suppose I'm happy that he is no longer president if he has strong opinions on topics he is clueless.

I don't think they're strong opinions

People often get asked to weigh in on things, and then news headlines run with the responses. Sure everyone could say "I don't know enough so I won't say anything", but that's a little unproductive.

I find it annoying especially when some news agency asks a loaded question, and then regardless of the response they have some story to run with