This seems like perfectly correct enforcement of that rule. Yes, it's a terrible rule – it always has been. I am baffled that someone would think this is any different from how the App Store has been managed for the last 18 years. Walled gardens gonna walled garden.
Apple's app store rules have never been compatible with devtools. It's kind of surprising to me that a Replit app existed on iOS at all; I would have expected that to be a nonstarter, and, given that a Replit app does somehow exist, I'm not surprised that they wound up unable to update.
This is a big part of why I don't use any iOS devices. It's possible to sort of buy your way out of the restrictions by paying for a developer subscription, but at the end of the day it's way too totalitarian.
Maybe I'm just not into the iOS ecosystem enough, but I find this article really hard to take seriously.
The entire discussion is predicated on the existence of "adaptive software", not defined here and with no arguments as to why that's different than extensions or software interpreter.
The article itself seems confused. If Replit has it's created apps being displayed within its app boundaries, how is that different than an app including an interpreter? The article states "you can't version and review adaptive software" but you sure can do that with an interpreter. The fact that you can create additional software doesn't mean the software you are using to do that isn't normal software.
Yes, you're right that interpreters also allow users to run code. And we could argue that Apple is simply being inconsistent in how it applies its policies. I think realistically the difference is in how popular these newer vibe coding apps are, but also the fact that they have a much broader scope of what can be generated.
With Pythonista or a Lua-scripted game, the reviewer can assess what's possible: this app can do everything Python-with-this-API-surface can do, and nothing more.
With LLM-driven generation, the set of possible behaviors isn't fixed. The same Replit app can produce totally different behaviors next month than it can today, without ever being resubmitted, based on model or system prompt updates.
That's what I meant with "you can't review adaptive software".
I think the main "issue" is that Apple won't be able to properly take their beholden 30% or whatever their fee is nowadays. A "wrapper" app has no real value by itself so they can't put a serious cost on it, but once it's on a user's device it can be used to create unlimited things of high value. Apple definitely doesn't want to miss out on taxing that value in their ecosystem, so it's easier to just block it and force devs to stick to apps with a appraisable value.
The set of behaviors is certainly fixed or constrained with an LLM exactly the same way as an interpreter- if neither one has an escape from the pre-programmed APIs included in the app, they can’t escape regardless of what they do. Apple allows interpreters that run entered or downloaded code like programmable calculators and even more powerful environments. There’s nothing special about an LLM creating an app.
One of the reasons why these kind of software is not allowed is that it opens up a new class of social engineering attacks (install this, run this code) against users that have normally placed great trust in their software.
Low-code apps existed. Terminals existed. Apple rejected them.
LLM apps exist. Apple allows them, because they render text, pictures, and video, but they don't run arbitrary code.
Running arbitrary code is flatly forbidden, because users can't reason about them. I see absolutely no evidence that software is moving away from versions, any more than it was when apps could first search the internet, render recommendations, or deliver messages.
Something not really discussed but almost certainly a factor in Apple's reasoning is the blowback if someone vibe codes some type of app which creates a scandal, and it's reported as if it's an app available on the App Store.
Look at the recent controversy with Grok on X as an example. Imagine BBC runs a story with the headline "iPhone App does {very_illegal_thing}", or "iPhone users can now do {extremely_morally_objectional_thing}", and then idiot governments start trying to regulate or issue fines. One reason for the rule is so their review process prevents these things from happening.
No, not strange, selective enforcement is what's strange.
In London, it's illegal to shake rugs in the street. If police actually starts prosecuting people for that, and not all people but just bald ones, it's natural that people won't be happy and start asking questions about the anti-bald bias.
> No, not strange, selective enforcement is what's strange.
It's not strange at all, and in fact it's unavoidble. Rules and laws have always been enforced with discretion because of external considerations like limited resources, proportionality, mercy, prioritization, etc.
The argument in the post isn't that the enforcement is unfair, more that the rule might not make sense much longer now that software can write itself. Rule was written for a world where the artifact reviewed and the artifact running were the same thing. That assumption is breaking, and not just for vibe-coding apps.
The argument is silly, dev tools that allow you to run code were never allowed. There is no selective enforcement here and nothing has changed doesn’t matter if the code was written by a human or not.
Apple doesn't want software that can write itself available on its app store. I don't want software that can write itself running on my device. So for once Apple's priorities and the consumer's align.
Sudden decision? This looks completely in-line with what Apple has been doing since the launch of the iPhone.
I find the article most charitable to the idea that AI generated software is a different category than human generated software. It's merely a dev tool.
Not AI generated software -- DYNAMICALLY generated software, like at run time and ongoing. Even in-app directed by the user. This is not a thing that existed before, a degree of customizability well beyond letting the user pick a color scheme or from one of a few layouts or default start screens.
I don't know how good of an idea it would be, product-wise, to give programming level flexibility. I am reminded of greasemonkey scripts, but written in english maybe. Maybe it could be awesome. But Apple is saying "nope. Not interested in exploring this with you. BYE"
> Not AI generated software -- DYNAMICALLY generated software, like at run time and ongoing.
They don't allow general purpose software interpreters on their platform for the purpose of distributing apps. This is not new. AI adds nothing meaningful here.
Thank god there is at least one major platform preventing applications from doing this. I get in a perfect world this would be great, but we live in a world where we use apps built by corporations which do not have the user's interests at heart and their apps that we begrudgingly use are treated as practically antagonistic.
At some point in life, you realize that some discussions, some arguments, are entirely wastes of energy and entire premises can be thrown away with simpler discussions and arguments.
To this one, I say, who cares? Don’t publish on platforms where you can’t control your own intellectual property.
Why use the App Store at all? It only serves to benefit Apple, and the vast majority of developers are simply making $100/yr payments to use their own custom software.
For the largest companies who ship apps on iOS and Android because they have more money than sense and can afford to waste countless engineering hours letting barely qualified, fractionally compensated people say yes or no, I say let them.
For the rest of us who are better managers, let’s own our own release process, and if that means building a website or a web app instead, go do it.
Whatever you’re shipping to a phone isn’t for professionals anyway.
Maybe, but we pay notably more to outsource than we were previously paying for in-house dev; it works better at our scale - our client apps probably merit like 1 FTE developer, and it's easier to get appropriate expertise for mobile app development from a third party that specializes in that rather than trying to maintain it in-house, or by spreading the pain of Apple provisioning profiles and app store review, etc. among a team of non-mobile devs.
This seems like perfectly correct enforcement of that rule. Yes, it's a terrible rule – it always has been. I am baffled that someone would think this is any different from how the App Store has been managed for the last 18 years. Walled gardens gonna walled garden.
Apple's app store rules have never been compatible with devtools. It's kind of surprising to me that a Replit app existed on iOS at all; I would have expected that to be a nonstarter, and, given that a Replit app does somehow exist, I'm not surprised that they wound up unable to update.
This is a big part of why I don't use any iOS devices. It's possible to sort of buy your way out of the restrictions by paying for a developer subscription, but at the end of the day it's way too totalitarian.
Apple enforces their rules rather arbitrarily. Wouldn't emulators also fit the definition of self blunded apps?
Anyway, I am glad I have never owned any Apple products in my life.
They can screw themselves with their tight grip on user freedom.
Maybe I'm just not into the iOS ecosystem enough, but I find this article really hard to take seriously.
The entire discussion is predicated on the existence of "adaptive software", not defined here and with no arguments as to why that's different than extensions or software interpreter.
The article itself seems confused. If Replit has it's created apps being displayed within its app boundaries, how is that different than an app including an interpreter? The article states "you can't version and review adaptive software" but you sure can do that with an interpreter. The fact that you can create additional software doesn't mean the software you are using to do that isn't normal software.
Yes, you're right that interpreters also allow users to run code. And we could argue that Apple is simply being inconsistent in how it applies its policies. I think realistically the difference is in how popular these newer vibe coding apps are, but also the fact that they have a much broader scope of what can be generated.
With Pythonista or a Lua-scripted game, the reviewer can assess what's possible: this app can do everything Python-with-this-API-surface can do, and nothing more.
With LLM-driven generation, the set of possible behaviors isn't fixed. The same Replit app can produce totally different behaviors next month than it can today, without ever being resubmitted, based on model or system prompt updates.
That's what I meant with "you can't review adaptive software".
I think the main "issue" is that Apple won't be able to properly take their beholden 30% or whatever their fee is nowadays. A "wrapper" app has no real value by itself so they can't put a serious cost on it, but once it's on a user's device it can be used to create unlimited things of high value. Apple definitely doesn't want to miss out on taxing that value in their ecosystem, so it's easier to just block it and force devs to stick to apps with a appraisable value.
That too for sure
The set of behaviors is certainly fixed or constrained with an LLM exactly the same way as an interpreter- if neither one has an escape from the pre-programmed APIs included in the app, they can’t escape regardless of what they do. Apple allows interpreters that run entered or downloaded code like programmable calculators and even more powerful environments. There’s nothing special about an LLM creating an app.
One of the reasons why these kind of software is not allowed is that it opens up a new class of social engineering attacks (install this, run this code) against users that have normally placed great trust in their software.
This is ... not new at all?
App-bundling apps existed. Apple rejected them.
Low-code apps existed. Terminals existed. Apple rejected them.
LLM apps exist. Apple allows them, because they render text, pictures, and video, but they don't run arbitrary code.
Running arbitrary code is flatly forbidden, because users can't reason about them. I see absolutely no evidence that software is moving away from versions, any more than it was when apps could first search the internet, render recommendations, or deliver messages.
Really? Apple lets you run custom code with network connectivity and even payments now. So how does that work? There's:
https://apps.apple.com/us/app/roblox/id431946152
And to:
https://apps.apple.com/us/app/minecraft-play-with-friends/id...
And to IDEs:
https://apps.apple.com/us/app/go-programming-compiler/id1503...
https://apps.apple.com/us/app/python-coding/id6738411328
https://apps.apple.com/us/app/pythonista-3/id1085978097
https://apps.apple.com/us/app/python-editor-app/id1672453872
And then of course there's emulators, including:
https://apps.apple.com/us/app/utm-se-retro-pc-emulator/id156...
The famous one:
https://apps.apple.com/us/app/delta-game-emulator/id10485246...
I see no reason those emulators wouldn't fall under the Blunded App rule, they "execute code" (interpret instructions).
It's just Apple enforcing their rules when they see fit...
Something not really discussed but almost certainly a factor in Apple's reasoning is the blowback if someone vibe codes some type of app which creates a scandal, and it's reported as if it's an app available on the App Store.
Look at the recent controversy with Grok on X as an example. Imagine BBC runs a story with the headline "iPhone App does {very_illegal_thing}", or "iPhone users can now do {extremely_morally_objectional_thing}", and then idiot governments start trying to regulate or issue fines. One reason for the rule is so their review process prevents these things from happening.
[dead]
Rules always existed and were clear. Outrage that they are being enforced seems strange.
No, not strange, selective enforcement is what's strange.
In London, it's illegal to shake rugs in the street. If police actually starts prosecuting people for that, and not all people but just bald ones, it's natural that people won't be happy and start asking questions about the anti-bald bias.
> No, not strange, selective enforcement is what's strange.
It's not strange at all, and in fact it's unavoidble. Rules and laws have always been enforced with discretion because of external considerations like limited resources, proportionality, mercy, prioritization, etc.
The argument in the post isn't that the enforcement is unfair, more that the rule might not make sense much longer now that software can write itself. Rule was written for a world where the artifact reviewed and the artifact running were the same thing. That assumption is breaking, and not just for vibe-coding apps.
The argument is silly, dev tools that allow you to run code were never allowed. There is no selective enforcement here and nothing has changed doesn’t matter if the code was written by a human or not.
Apple doesn't want software that can write itself available on its app store. I don't want software that can write itself running on my device. So for once Apple's priorities and the consumer's align.
The outrage isn't about the rule. It's about the sudden decision to choose bureaucracy over common sense.
Sudden decision? This looks completely in-line with what Apple has been doing since the launch of the iPhone.
I find the article most charitable to the idea that AI generated software is a different category than human generated software. It's merely a dev tool.
Not AI generated software -- DYNAMICALLY generated software, like at run time and ongoing. Even in-app directed by the user. This is not a thing that existed before, a degree of customizability well beyond letting the user pick a color scheme or from one of a few layouts or default start screens.
I don't know how good of an idea it would be, product-wise, to give programming level flexibility. I am reminded of greasemonkey scripts, but written in english maybe. Maybe it could be awesome. But Apple is saying "nope. Not interested in exploring this with you. BYE"
> Not AI generated software -- DYNAMICALLY generated software, like at run time and ongoing.
They don't allow general purpose software interpreters on their platform for the purpose of distributing apps. This is not new. AI adds nothing meaningful here.
Maybe it could, but I don't want it on my phone, and I like the fact that I have at least some confidence that an app I install isn't that.
If I don't want Instagram on my phone should Apple ban Instagram?
Thank god there is at least one major platform preventing applications from doing this. I get in a perfect world this would be great, but we live in a world where we use apps built by corporations which do not have the user's interests at heart and their apps that we begrudgingly use are treated as practically antagonistic.
Good.
It's a shame we let ourselves get to this point where creativity is limited by the walled gardens we've almost fully accepted as "worth it".
The people who opt into walled gardens don't want to create, they want to consume. Those who create want unfettered access.
I like both. I keep a Linux system (or 4) around for more creative stuff, but I use a Mac as my default machine.
iPads are great for specific things, as are phones, as are laptops/desktops.
Strange that the Apple walked garden creates more apps than the “free” Android world than.
Plenty of creators in the walled garden.
At some point in life, you realize that some discussions, some arguments, are entirely wastes of energy and entire premises can be thrown away with simpler discussions and arguments.
To this one, I say, who cares? Don’t publish on platforms where you can’t control your own intellectual property.
Why use the App Store at all? It only serves to benefit Apple, and the vast majority of developers are simply making $100/yr payments to use their own custom software.
For the largest companies who ship apps on iOS and Android because they have more money than sense and can afford to waste countless engineering hours letting barely qualified, fractionally compensated people say yes or no, I say let them.
For the rest of us who are better managers, let’s own our own release process, and if that means building a website or a web app instead, go do it.
Whatever you’re shipping to a phone isn’t for professionals anyway.
> Whatever you’re shipping to a phone isn’t for professionals anyway.
That's a bold statement.
I hate developing for iPhone, but I don't have a choice because 85% of the users of our B2B app are on iOS.
Have you tried using a web app? A well made PWA app can work well, and most iPhone users don't care about memory or lag.
> I hate developing for iPhone, but I don't have a choice because 85% of the users of our B2B app are on iOS.
I mean, you could outsource the mobile app building.
My company does basically our entire B2B stack and infrastructure in-house except for the mobile apps.
That just indicates you don’t really care about your customers mobile experience and they will notice eventually.
Maybe, but we pay notably more to outsource than we were previously paying for in-house dev; it works better at our scale - our client apps probably merit like 1 FTE developer, and it's easier to get appropriate expertise for mobile app development from a third party that specializes in that rather than trying to maintain it in-house, or by spreading the pain of Apple provisioning profiles and app store review, etc. among a team of non-mobile devs.
Aren't ipads the new clipboards?
I've never seen someone fling their clipboard against the wall so hard it shatters. Probably clipboard developers are just a better breed.
The go to move with clipboards is to break them in half across your knee/leg.