throwaway4aday

I'm going to tap the sign again:

  - [X] Text
  - [X] Images
  - [X] Audio
  - [ ] Videos (in progress)
  - [ ] 3D Meshes and Textures (in progress)
  - [ ] Genetics (in progress)
  - [ ] Physics Simulation (in progress)
  - [ ] Mathematics
  - [ ] Logic and Algorithms aka Planning and Optimization
  - [ ] Reasoning
  - [ ] Emotion
  - [ ] Consciousness

We still have a lot of data to crunch but it's not nearly enough so we're also going to have to collect and generate a lot more of it. Some of these items require data that we don't even know how to collect yet. Barring some kind of disastrous event, draconian regulation, or politically/culturally motivated demonization of ML I don't see GPU demand dropping any time soon.

That's great to hear! It's uses like this that really make Udio and other tools shine. Even just making up silly songs about things that happen in your life is fun or doing them as a gift like you did is always nice. It's also great to have the option to add music to other projects.

Fair enough but I'm not sure you're even going to realize if you're listening to AI generated music or not. One way of using these tools is to take lyrics and create several different melodies and vocal styles. An artist or a professional songwriter could do this and then either record their own version or pay musicians to perform it. That could be any combination of simply re-recording the vocal track, replacing or adding instrument tracks, making small modifications to some of the AI generated tracks, etc. The song can then be released under the name of the singer who can also go on tour in the flesh. You might also just come across a 100% AI song on a streaming platform and enjoy it and add it to a playlist. Who vets all of the music they listen to anyways? and if the producer manages the web presence of the "band" and provides a website it would withstand a cursory search. You'd have to look closely to determine that there are no human band members other than the producer. For the types of electronic music that aren't usually performed live and are solely attributed to one artist it might be impossible to tell. The line would be especially blurry there anyways due to the current extensive use of samples and non-AI automation.

There are a lot more fuzzy edges too, you can use AI tools to "autotune" your own voice into a completely different one. You can tap out a quick melody on a keyboard and then extend, embellish and transform it into a full song. You could even do the full song yourself first and then remix it using AI.

The point I agree on would be that one-click hits are going to be few and far between for a while at least. If no effort is put into selecting the best then it's really just random chance. I'd be willing to bet that there will be an indie smash hit song created by a single person who doesn't perform any of the vocals or instruments within a year though. It'll get no play time on anything controlled by the industry titans but people will be streaming it regardless.

It sounds like you're arguing with yourself. You provide exactly the reasons why generative AI isn't going to take us to a "place where human musicians can't earn a living". It's my understanding that most small bands primarily earn their money from live performances and merchandise, gen AI isn't going to compete with them there, if anything it'll make it much easier for them to create their own merch or at least the initial designs for it.

AI generated music is more of a threat to the current state of the recording industry. If I can create exactly the album or playlist that I want using AI then why should I pay a record label for a recording that they're going to take 90% of the retail price from? The playlist I listen to while I'm working out or driving is not competing with live band performances, I'm still going to go to a show if there's a band playing that I like.

Have you tried the latest model? It's night and day.

Edit:

There's obviously still skill involved in creating a good song, it's not like you can just click one button and get a perfect hit. I outlined the simplest process in my first comment and specifically said you could create a "listenable" song, it's not going to be great but it probably rivals some of the slop you often hear on the radio. If you're a skilled music producer you can absolutely create something good especially now with access to the stemmed components of the songs. It's going to be a half manual process where you first generate enough to capture the feeling of the song and then download and make edits or add samples, upload and extend or remix and repeat.

If you're looking for links and don't care to peruse the trending section they have several samples on the announcement page https://www.udio.com/blog/introducing-v1-5

Here's an experiment you can try, go to https://www.udio.com/home and grab a free account which comes with more than enough credits to do this. Use a free chat LLM like Claude 3.5 Sonnet or ChatGPT 4o to workshop some lyrics that you like, just try a few generations and ask it to rewrite parts you don't like until you have something that you don't find too cringe. Then go back over to Udio, go to the create tab turn on the Manual Mode toggle and type in only 3 or 4 comma separated tags that describe the genre you like keep them very basic like Progressive Rock, Hip Hop, 1995, Male Vocalist or whatever you don't need to combine genres these are just examples of tags. Then under the Lyrics section choose Custom and paste in just the chorus or a single verse from the lyrics you generated and then click Create. It'll create two samples for you to listen to, if you don't like either of them then just click Create again to get another two but normally it doesn't take too many tries to get something that sounds pretty good. After you have one you like then click on the ... menu next to the song title and click Extend, you can add sections before or after and you just have to add the corresponding verse from the lyrics you generated or choose Instrumental if you want a guitar solo or something. You'll wind up with something pretty good if you really listen to each sample and choose the best one.

Music generation is one of the easiest ways to "spook the normies" since most people are completely unaware of the current SOTA. Anyone with a good ear and access to these tools can create a listenable song that sounds like it's been professionally produced. Anyone with a good ear and competence with a DAW and these tools can produce a high quality song. Someone who is already a professional can create incredible results in a fraction of the time it would normally take with zero budget.

One of the main limitations of generative AI at the moment is the interface, Udio's could certainly be improved but I think they have something good here with the extend feature allowing you to steer the creation. Developing the key UI features that allow you to control the inputs to generative models is an area where huge advancements can be made that can dramatically improve the quality of the generated output. We've only just scratched the surface here and even if the technology has reached its current limits, which I strongly believe it hasn't since there are a lot of things that have been shown to work but haven't been productized yet, we could still see steady month over month improvements based on better tooling built around them alone.

Text generation has gone from markov chain babblers to indistinguishable from human written.

Image generation has gone from acid trip uncanny valley to photorealistic.

Audio generation has gone from 1930's AM radio quality to crystal clear.

Video generation is currently in fugue dream state but is rapidly improving.

3D is early stages.

???? is next but I'm guessing it'll be things like CAD STL models, electronic circuits, and other physics based modelling outputs.

The ride's not over yet.

Average lifespan has increased, meaning more people survive early deaths due to disease and injury that we can now prevent, but as individuals we haven't broken through any maximum age barriers.

That said, sun bathing is a modern phenomenon. Our ancestors knew enough to assess that painful sunburns were bad and mitigated their exposure by wearing large hats and clothing with long sleeves. I'm sure even primitive peoples would cover themselves when necessary. If you spend a lot of time working outside you also develop a very dark tan which helps quite a bit. Modern people seem to suffer mostly from spending so much time indoors that even moderate sun exposure on the weekend produces a damaging sunburn.

Manufactured products only get cheaper by scaling and improving the manufacturing process. You should know this because it's the reason why silicon photovoltaics have gotten cheaper after starting at some rather eye watering prices. Land is a finite resource with an extremely complex system of regulations, laws, ownership, uses and speculation built up over 10,000 years of history and any conversion of land for one purpose necessarily excludes it for many other purposes so there is massive competition over prime real-estate and many un-intended future consequences resulting from development.

Not worrying about land use when it comes to power generation is very, very much like not worrying about carbon emissions when it comes to power generation/transport/manufacturing "The atmosphere is enormous! Why should we worry about emissions when our current demand for power and transport is so tiny?"

You have to look at it as a contractor. If you tell a contractor to "build me X" then you might get anything back with a high probability of getting something common but outdated. You have to write a specification for it with all of the constraints and preferences you have. Works well if you know the domain, if you're using it for something you don't know much about then you have to do more of the legwork yourself but at least it will give you a starting point that can inform your research.

Not to pick on you but this general type of response to these AI threads seems to be coalescing around something that looks like a common cause. The thing that tips your comment into that bucket is the "serious tasks" phrasing. Trying to use current LLMs for either extremely complex work involving many interdependent parts or for very specialized domains where you're likely contributing something unique or to any other form of "serious task" you can think of generally doesn't work out. If all you do all day long are serious tasks like that then congrats you've found yourself a fulfilling and interesting role in life. Unfortunately, the majority of other people have to spend 80 to 90 percent of their day dealing with mind numbing procedural work like generating reports no one reads and figuring out problems that end up being user error. Fortunately, lots of this boring common work has been solved six ways from Sunday and so we can lean on these LLMs to bootstrap our n+1th solution that works in our org with our tech stack and uses our manager's favourite document format/reporting tool. That's where the other use cases mentioned in the article come in, well that and side projects or learning X in Y minutes.

It's surprisingly good at helping diagnose car problems as well.

You can only learn how to draw owls by drawing owls, repeatedly. The more owls you draw, the better you will get so it helps to draw quickly and often.

Well, that's because it is how you draw an owl. You start with basic outlines to define size, pose and proportion, then you divide those outlined areas into smaller sections that locate important details like eyes, beak, ears, wings, etc. then you add rough detail to those parts which further breaks it down and you keep iterating like that until you fill in all of the fine details.

You can see the process here https://youtu.be/UGVOE2eHl0I

Ironically, to meet your definition of free you would have to violate the definition of community provided in the article.

The experiment in the paper is not well designed. They are repeatedly fine tuning the model and replacing the entire data set each time with a noisier version. That's just not how the world works and is literally the most naive approach you could take. They should have attempted to increase the size of the training set using output from the model combined with human editing and input and figured out a good evaluation strategy. That would have at least approached reality and may have produced useful knowledge. The fact still remains that the paper is hopelessly far behind the sota and almost entirely divorced from the processes it intends to make claims about.

The same problem exists if you blindly trust any source without verifying it. There is a huge amount of endlessly recycled incorrect blog spam out there for all domains. Not only that but this problem has always existed for second hand information so it's not like we were even starting from some pristine state of perfect truthfulness. We have the tools we need to deal with the situation and they were developed hundreds of years ago. Empiricism being chief among them. Nullius in verba[0]

[0] https://en.wikipedia.org/wiki/Nullius_in_verba

It's an interesting but fundamentally flawed idea. My suggestion would be to play with it and have fun but don't bet the house on it.

If you're curious what the flaw is, think Empiricism vs Rationalism.

lawyer: "don't worry, it's only attempted murder. I'm sure the jury will see things your way."

Why were the deaths at all the other rowdy protests downplayed?

> They should get to influence national decisions based on their population

That is exactly mob rule. Why should any group of people be allowed to enforce its will on another group simply because they are more numerous? If there are no systems put into place to limit and balance the will of the majority then they are perfectly capable of running amok. How do you think things like lynchings and pogroms happen? A majority decides the minority is to blame for some evil and they happily ignore laws and morals because they have come to a unanimous decision. The same thing can happen at national scale.

Imagine if you applied your population rule to the U.N., what a farce that would be.

A simple majority vote only works in relatively small groups where all of the participants are on roughly equal standing and even then it's a good idea to have some secondary authority re-evaluate the decision and ensure everyone is acting in good faith and not simply forming temporary alliances in order to rob some weaker group and divide the spoils.

This is really just wasting 20 minutes, it's much better to have a clear agenda which is basically in the form of bullet point notes that most people would make outlining the important topics and questions. Then just go through the agenda providing a brief elaboration on each topic and take questions, get agreement, write down the conclusion and move on. If the meeting is purely informational then in most cases you shouldn't even have a meeting.

If the bot isn't abusing the service and is operating at a speed roughly equivalent to a human and your intent is to distribute the service freely or the bot is paying a fee for access then what's the problem? Anything I can think of would also be abuse of the terms of service or code of conduct and falls in the same category as regular meat people doing the same things.

What a nothing-burger, Figma says they didn't do any custom training or fine tuning and they think the look a like problem is due to a design system they commissioned. They don't explicitly say the design system was created by a human but odds are... Also, the thing that was "ripped off" is the weather app. Oh, you mean the app that looks nearly identical on every platform and website ever? Wow, I'm shocked.

This is awesome! I've done a lot of little projects exploring the use of graphs with LLMs and it's great to see that this approach really pays off. Stupid me for trying to prematurely optimize when the solution is just prompt engineering and burning a bunch of tokens on multiple passes. Going to give this a try and see if my jaw drops. If it's as good as it looks then I'll have to put in the work to get it out of Python-land.

Faiss is for similarity search over vectors via k-NN. GraphRAG is, well, a graph. More precisely, GraphRAG has more in common with old school knowledge graph techniques involving named entity extraction and the various forms of black magic used to identify relationships between entities. If you remember RDF and the semantic web it's sort of along those lines. One of the uses of Faiss is in a k-NN graph but the edges between nodes in that graph are (similarity) distance based.

Looking at an example prompt from GraphRAG will make things clear https://github.com/microsoft/graphrag/blob/main/graphrag/pro...

especially these lines:

Return output in English as a single list of all the entities and relationships identified in steps 1 and 2.

Format each relationship as a JSON entry with the following format:

{{"source": <source_entity>, "target": <target_entity>, "relationship": <relationship_description>, "relationship_strength": <relationship_strength>}}

I'd just try setting the env vars and see what happens, if it breaks you'll have to read the code though.

https://microsoft.github.io/graphrag/posts/config/overview/

When I consider my own LLM workflow the amount of time reading/listening/thinking outweighs the amount of time spent typing/speaking. If that's any indication of how a fully fledged conversational workflow would work then I think open plan offices wouldn't be a lot louder than they currently are. Depending on how quickly agentic LLMs are developed I'm not even sure we will be using offices the same way we are now. We might only need to meet or checkin with our coworkers and our LLM agents every few hours or once a day or maybe even longer in order to realign and check on results. Maybe we'll get occasional messages asking us to confirm something or provide clarification, I could honestly see most knowledge work evaporating and leaving behind only high level coordination, research and ideation.

Before that, I'm certain we'll all be spending a lot more time reviewing work, trying out prototypes and tweaking prompts or specifications than we do typing or talking.

Request issued for additional platinum-iridium rich asteroids to be delivered to Earth orbit.

Life is pretty good at adapting. If we did a lot of selective breeding and genetic manipulation ahead of time and then made a sustained effort to repeatedly seed the more temperate locations on Mars we might eventually see something survive. I do think it's a bit early for that stage of terraforming. We really should experiment with smashing asteroids into Mars to see if we can kick up enough CO2 to make a difference first. I also like the idea of repeatedly impacting a single crater in order to excavate a very deep and wide hole to trap atmospheric gasses and create enough pressure near the bottom for a more human friendly environment. It'd have the added benefit of depositing a bunch of metals and other elements in a convenient location where they could be mined and smelted.

Ignoring the snark. This will change as technology is adopted, go back 40 years (or even less) and a person walking around staring at a little black rectangle would have been perceived as weird and anti-social. We used to make fun of people talking on the phone via bluetooth headsets and now everyone does it with AirPods or whatever.

If you've got the technology to enable you to seamlessly transition from working in your home to working while sitting outside at a cafe to working while sitting on a blanket under a tree in the park to working wherever you feel like it then there will be enough brave people that say "fuck what other people think" and just do it so they can enjoy being active and getting fresh air and eventually more and more people will join them. Eventually we'll reach the point where sitting inside at a desk for 8-12 hours will be the weird thing.