Loopy Pro: Create music, your way.
What is Loopy Pro? — Loopy Pro is a powerful, flexible, and intuitive live looper, sampler, clip launcher and DAW for iPhone and iPad. At its core, it allows you to record and layer sounds in real-time to create complex musical arrangements. But it doesn’t stop there—Loopy Pro offers advanced tools to customize your workflow, build dynamic performance setups, and create a seamless connection between instruments, effects, and external gear.
Use it for live looping, sequencing, arranging, mixing, and much more. Whether you're a live performer, a producer, or just experimenting with sound, Loopy Pro helps you take control of your creative process.
Download on the App StoreLoopy Pro is your all-in-one musical toolkit. Try it for free today.
Comments
Sentience/consciousness is a completely different issue from intelligence. They are orthogonal, in computer geekspeak. Consciousness seems to go along with biological organisms that have emotions. Clearly AI does not function in that way.
Well... look what I managed to find during my coffee break. I do like a good quote about code
When you actually look at the [generated] code, sometimes I get a little bit of a heart attack because it’s not super amazing code necessarily all the time. It’s very bloaty, there’s a lot of copy-paste, there are awkward abstractions that are brittle. It works, but it’s really just gross.
Andrej Karpathy, co-founder of OpenAI speaking at AI Ascent April 29, 2026
What AI is actually good for: memes
If you think that's bad, take a look at the source code for vibe coded websites.
As someone who has to host, maintain, and keep working the websites I build, working on that stuff would have me out of business in months.
Who better to turn to than ChatGPT for an explication of what Karpathy actually means by "there's a lot of copy-paste":
For the full ChatGPT reaction to that quote, you can read the response here:
https://chatgpt.com/share/6a0c50b3-ba94-83ab-b4b6-a04b5e17cafe
.
@dendy - I was specifically referring to the data used to train these models.
Understands the logic? I beg to differ...
AI agents do not "understand" the world the way humans do. Instead of actual comprehension, they rely on advanced pattern recognition, statistical probability, and logic. They simulate understanding by predicting the next most logical word or action based on vast amounts of data.
And that is not how I work
@hes - cool that you did that.
I'd say that response (well worth reading) was fair, and also my impression of the AI generated code I've seen in repos.
Hmm. I'm not so sure that's not how you work.
For anyone who cares, and turning again to ChatGPT, here's an explanation of what might be different and what might be similar between human and AI thinking and understanding:
https://chatgpt.com/share/6a0c585b-d1d4-83a5-8a6b-bf94133b83ea
@hes, granted there are similarities in the process. Otherwise it probably wouldn't work, right?
But at the risk of splitting hairs further, and the context is now vague because the post I was responding to has been, um... "re-factored" ?
Anyhoo, I was zoning in on the way most humans write code being very different to how an AI agent generates code. That said, I'm sure many software engineers just copy and paste from public repos and stack overflow, then tweak it a bit, so they're behaving more like AI agents in terms of process.
If you're still implying that AI copies and pastes from public code, then you're still way off. AI has indeed been trained on lots of publicly available code. In that code, AI has absorbed hundreds, thousands, or millions of instances of specific patterns, and has internalized those patterns. (Same as you and me?). It does not copy some specific instance. It is trained on many, many instances, each of which will influence how the relevant internalized pattern is represented, and those internalized patterns are what the AI uses when it writes code.
So AI is not generally copy-and-pasting from public repos and stack overflow like an inexperienced developer might. The main thing that AI shares with an inexperienced developer is an inability to grasp high level architecture on its own and more focus on how things work locally, which results in less elegant, more repetitive, and more brittle code.
Having said all that, I wish @dendy would weigh in again, because when you vibe code with an AI agent iteratively the duo becomes a form of pair programming, where the human can prompt the AI agent in ways that increase architectural awareness, and the end result can be better than either one writing code alone.
Also, AI's current deficiencies in understanding higher level architecture are, as I understand it, the result of its limited awareness of "context". Improvements in "context" are a primary focus of current development and will result in AI becoming better at higher level architectural understanding over time.
@hes - OK, yes, you got me,
I know it doesn't really simply copy and paste. Well... probably doesn't most of the time, and I'm aware that a very sophisticated multi-step process is at play here.
I agree with everything else you say.
It's just from the code I've seen, and the earlier quote from Andrej Karpathy kinda backs this up, is that sometimes, the generated code looks very much like that, and I quote, is "just gross". Is that the same as "slop" ?
And my over-arching concern amid all of this vibe coding related discussion basically boils down to:
I suspect there'll be a bit of, well, it works, so... I don't need to do anything, right?
But, if you're proactively reviewing the code, like an experienced developer would review another developer's pull request with suggestions to re-factor, etc., then great!
In a previous life, I worked at a place where pair programming was kinda forced on all the dev teams. I quite liked it because I liked the people I was paired with.
Sadly after a few months it was abandoned due to (a) numerous personaility clashes, and (b) a particularly memorable actual physical clash in the middle of the office... I didn't know backend developers had so much pent-up rage. Must be having to type all that shouty uppercase SQL...
+1
[wall of text trimmed on the topic of incorrectly postulated LLM overfitting as frequently used straw man]
PS. In case you're curious about the altercation, it apparently started with one of the guys snatching the keyboard away while the other was typing and it just sort of escalated from there.
Probably not something a coding agent would do, so there's something positive I can add.
Having worked with / taken over code from copypasta chefs, de-duplication is hard work. Pure horror is widespread duplication of domain logic, because you’ll always miss one case when time to update it…
Latest news is that vibe coding prices are going up by an order of magnitude, which kind of shifts the cost balance vs junior devs?
Sorry that i deleted my post but i decided I doesn't want to try to convince you
I am doing this mistake again and again and I should just stop.
I totally respect your approach and your work (and results of your work, mad respect man !) - so i think .. this discussion is just waste of time .. really, no bad feelings, just pure respect. Everybody is free to choose his path. As soo nas your methond works for you there is zero reason to change it, that is fact.
Yes, this is my personal experience.
@dendy - thanks, and I'm feeling the same way, so respect to you also.
Plus none of my criticism of the way in which some developers are using AI applies to you, i.e. you seem to know what you're doing
and more importantly, it's working for you.
My gut tells me it probably wouldn't work for me, for all sorts of reasons, so nothing more to add here other than apologise for the noise.
@Rob_Jackson_Music You crack me up, Rob! 🤣🤣🤣
Rob's got quite the knack with italics and bold, I'll give him that!
Hopefully, we can all have a chuckle at this.
In case you don't know albertatech, she's a very smart and funny Software Engineer at Google (that uses AI - of course!) but has a little side-hustle posting YT videos about tech life
https://youtube.com/shorts/ql56K3sveqo?si=IVbtX9toN1BjgBlH
To add to your point.
It is shocking to me how many people, including technical people that should know better, do not understand that the appearance of “thinking” is not the same as thinking.
It is a powerful tool. But it isn’t thinking in the human sense.
LLMs tokenize text (very cleverly) and do sophisticated statistical analysis to weigh the probability of which token comes next…,and that only looks like thinking if trained on an unimaginably large data set. Much much much much larger than humans need to achieve competence.
Human thinking and creativity does not require anywhere near the amount of data and training that an LLM needs. When I say nowhere near, the scale of difference is off by orders and orders and orders….. of magnitude.
Mozart was composing worthwhile and rapidly evolving music at a young age. The amount of data he was exposed to was (by LLM standards) so small that an untrained LLM would not be able to generate consistently engaging music if trained on the same data. Recorded music didn’t exist. The data input was only what he could hear live.
The same is true of all of us. We learn fluency in our native tongue and the ability to solve problems with a tiny fraction of what an LLM needs to be trained on.
That by itself should make it obvious that human thinking is a very different process from what AIs do.
This not a knock on the useful things they can do. But I think the anthropomorphizing of LLMs is off base.
@espiegel123 said: “Human thinking and creativity does not require anywhere near the amount of data and training that an LLM needs. When I say nowhere near, the scale of difference is off by orders and orders and orders….. of magnitude.
Mozart was composing worthwhile and rapidly evolving music at a young age. The amount of data he was exposed to was (by LLM standards) so small that an untrained LLM would not be able to generate consistently engaging music if trained on the same data. Recorded music didn’t exist. The data input was only what he could hear live.
The same is true of all of us. We learn fluency in our native tongue and the ability to solve problems with a tiny fraction of what an LLM needs to be trained on.
That by itself should make it obvious that human thinking is a very different process from what AIs do.”
without xai(explained AI) we don't know the weights (black box neural nets) used to understand the reasoning of AI outputs.
Once we do crack it, it'll be too late for any course correction.
It's important to have some perspective on this, I think. An AI gets trained starting out as a blank slate. Human minds are not like this, humans are born with a brain structure that has already been shaped by millions of years of evolution. Clearly there are differences. The correct conclusions to draw from this seem less obvious, IMO.
I think it's important to recognize that humans have a psychological desire to believe they're unique, and better than other beings. This is clearly displayed in religions, e.g., Christianity, where humans are made "in the image of God" are are not on the level of other animals, which are in almost all ways quite similar organisms. . . . Even in secular thought, humans are often credited with having a "soul", which is often denied to other animals. We humans like to believe we're special, and not just special, but better, like on a totally different plane of existence . . . .
LLM's are "born" with way more data and the ability to work with the data at hand immediately. Humans have filters that diminish and limit our abilities to work with the data at hand, even after a lifetime of learning.
A perfect example is a traumatic childhood of humans.. the phrase " it's all I knew" is a common utterance. Well, AI's do not have that excuse. It has immense knowledge at the moment of iteration.
.
The main reason our intelligence is different from AI is exactly because, like animals, its embodied. I don't think we're that special compared to other animals. But animal intelligence is, I think, fundamentally different from machine intelligence.
Yes, animal intelligence is in a biological organism and machine intelligence is in a digital computer. That's certainly a difference. If it is the neural net structure that is important, though, it's simply a similar structure in a different substrate. Of course, animals also have emotions and everything that goes with that, which machine intelligence lacks.
(Also, as an aside for the younger among us: I think today it's pretty widely acknowledged that animals other than humans have emotions. Dogs, cows, gorillas, rabbits, birds, etc. But fifty, sixty, seventy years ago this was quite a minority opinion. Although all these animals displayed behavior that was similar to humans, most people denied that they were "real" emotions. They were just a simulation of emotions, and the animals were actually just "automatons", beings that operated more like machines than like humans. How animals are regarded has changed quite a lot in a relatively short period of time. . .
[doing a little checking on this, I think maybe I should limit myself to saying that 'among scientists' it was a majority opinion; among regular people it might not have been a majority opinion, but was certainly a widely held one. "Animals don't have real emotions; they merely simulate emotions". Reminds me a bit of the common current thought that "AI don't really understand things; they merely simulate understanding." ])
To me that would just imply the word 'understanding' is faulty and needs to be upgraded or depricated.
Are we sure we don’t ingest as much data? Sure, we don’t download the entirety of human writing when we’re born but the amount of data we’re constantly being exposed to isn’t insignificant if you think about it from a sensor perspective. Atmospheric data, sounds, sights, smells, words, faces, etc. Our bodies are constantly receiving external input.