I bet it's pretty close to a Pareto distribution with 20% of the apps making 80% of the total App store revenue. You can guess which apps those are from Apple's top/editor's choice charts.
Do you even need the top chart when you open the app and every other day it's either a Scopely property or Monopoly Go?
It's wild to know that Freecash is somehow back in the App Store after [1] earlier this year and they're actively running aggressive marketing campaigns on TV again.
I think the app store will be in trouble in the future.
I think most non-gaming apps will be replaced by a single AI app that is like a super assistant to you. No need to book on Booking.com app if you can just have your assistant look for what you actually want.
Assistant booking a place for me is last thing I'd want, I want to see ratings, photos, overall look&feel of the place, not spin the wheel and get some random place booked
a) I won't trust booking's own assistant or "AI" and it will be garbage if/when they introduce it (despite fact that in order to handle it correcly it would need to have amount of information about me and my family and my trip that im not comfortable to provide them)
b) They will block 3rd party/independent assistant via CF or other measures - they want you to be on their property browsing on their terms upselling you features, not access via API/MCP
c) chat interface and general UI clunkiness related to AI and this kind of use case - when browsing booking I'm able to narrow to offers I want quickly, where accommodations are located on the map etc, lol, just thinking about doing all of that via chat UI and even more via voice interface makes me tired
I think the single AI app will be thoroughly enshittified. You ask for a hotel and the assistant operator is normally getting 20% of your payment as a kickback, so it shows you the highest price it thinks you'll pay, but 40% if it's a certain chain so it only shows that chain. Everything will be like this.
Am I one of very few people that believes LLM apps don't solve or replace a large number of apps.
Take all these basic lifestyle CRUD apps. Maybe it's a skill issue on my part, but whenever I've tried to use LLM apps over any sort of medium-to-long-term time scale for tracking something (ie. endurance running training), within about a month it begins hallucinating, recalling the wrong things, linking the wrong bits of data to other bits of data (dates, distances, heart rate data, workout paces etc), and it does all of this even when pointed to a basic spreadsheet or simple text file to keep it onboard with what happened where. It just seems to look at these files, and then look at it's own context and confuse all the details based on its own context.
Every other month, some brand new shiny model is released and it's exactly the same deal every time. I think the underlying technology simply can't deal with these sorts of goals. A whole range of apps for lifestyle tracking don't seem under threat, in my opinion, if this is the best its capable of.
Many other things I can't imagine wanting an AI to do for me, especially anything involving payments, bookings. There was a discussion the other day on here about AI being used to order takeout. Why would somebody want to delegate one of their last personal decisions to a LLM?
A good alternative for your example is to make it create a stateful artefact (Claude) / whatever OpenAI calls ChatGPT.
And there, it has a database of sorts, running god knows where on god knows what, and you have a single-user web app that you may or may not ever open and see (since your favorite AI agent can just read and update it for you anyway).
Today it’s still something the user has to suggest / request.
Tomorrow? If they put a bit of effort these models will reach for building these one-off ephemeral or long-lived hyper-custom apps the same way they now reach for a python sandbox for basic arithmetics and statistics.
Looking for a flat? Scheduled task + CRM like artefact for it to monitor, maintain, and update.
Looking for holidays? Your own personal shareable moodboard with destinations and what not.
Have a use case or task for which chat or editing text docs is a frustrating way of correcting or giving feedback? Have it build a dedicated, custom artefact just for that.
When you encounter a use case where LLMs limits show, make them build tools.
Your last line is actually very insightful. I was just looking at LLM from the chat app perspective.
I wonder how open these services will be to data storage if they do become a "universal app". If I used LLM to make an app for tracking training for endurance running, and I slapped an 800kb .fit file in there each day, the storage for all of these files would probably be fine, but without condensing each run down to a textual summary like Garmin/Strava currently does ("You ran 10km in 35:00, at a pace of 3:30/km. This improved your speed/{vague motivational LLM quote}"), wouldn't the context become enormous after a pretty short duration, causing huge costs? Genuinely curious.
What you're saying is true now, but given the existence of grok-bot and similar dots (et al.) there is now an amassing amount of training data for "personal assistants" -- so in a couple of generations time, you'll find frontier models will be extremely capable in these areas.
Harnesses have been developed for this, and its also fairly trivial to do large scale verifiable-rewards post-training for "personal assistant" style things. eg., invent a booking website, invent a PA GUI-aware harness, and reward the model when a successful booking takes place. Do that 1bn times.
The lifestyle monitoring thing is something I've been testing with ChatGPT, Gemini, Claude for the last year or two now, and it doesn't ever seem to get any better.
I may be completely out of the loop here, so forgive me if I'm missing details, but I always have and continue to believe that the coding prowess of LLM agents more-or-less comes down to the sheer abundance of data it has trained off of on the web, and that code itself is ideal for LLM by the way it is discrete and syntax-bound, same as the way LLM can write very well provided they've read more-or-less the same sequence of text before in abundance. For the speed advantages it provides in coding, and for the accuracy it has for well-established information, LLM is a very useful technology.
I just don't follow how they are going to do the same thing with such granular telemetry data, such as datasets from a day of nutrition or a workout. The values are continuous, minute, and are so parallel to noise in the way they are provided. Take running. If you dumped a .fit file from a Garmin or other watch with a series of heart rate values, paces, elevation levels, GPS co-ordinates, respiration rate, vertical oscillation etc, and you trained on a bunch of files like this, the values are so discrete that I can't understand how it would connect them and tokenize them across different people or even the same person's various workouts, with the high degree of factors (nutrition, climate, illness/injury state and severity) that can alter those values. It's no longer about just tokenizing text, but with advanced calculations.
I've seen second-hand through friends in running with apps like Runna that it produces plans that aren't suited for them at all. There's simply too many small variables that these things don't seem to be able to map. It gets caught up on summaries of runs. It can give general advice about things, but it doesn't have the level of detail or understanding necessary to give good guidance. Strava "AI" and Garmin "AI" just make a 2-3 line summary from the average data from the entire run or workout.
Perhaps this is something that is already being worked on, and I would be happy if it was, but I just can't personally see how LLM can work with such datasets in a detailed way, beyond the useless summaries Garmin and Strava give "You ran 10km in 35:00, with a pace of 3:30/km. This shows that your speed is improving". I would love an explanation/links on how LLM can get around this if anybody is willing to spend the time providing it. Happy to have my mind changed.
In summary: the LLM would most likely use a traditional ML method or python script to analyze the datasets and reduce the noise.
For example, I use LLMs to create 3d models. I don't tell them to recreate the 3d vertexes every time, they have a python script that programmatically creates the shapes, and with a freecad mcp, they get views in images, also they can create other scripts to check for collisions and other issues.
They can also get a histogram, discard noisy values, check outliers, etc, etc, using scripts or by running a traditional ML value to get later the refined data.
Opus 5.5 is quite decent currently, but expensive and not enough for many challenges, but I see the progress happening.
The big lift with LLMs, causing them to really explode in real-world capability, has been tool calling. Dumping data into the LLM only really works if the data fits very, very comfortably in the context window. The way the best harnesses work is that they provide the LLM a bundle of tools to manage context and memories on its own, and let the LLM use these tools to manage its understanding of the world.
The problem with these Garmin and Strava AI summaries is that most of them just point an LLM at a heap of data and ask them to offer a summary. They also usually use older models since the feature is not something particularly well curated. I do financial modeling with LLMs and use the LLMs to build/verify/iterate on models.
I don't think sites like booking are at risk, as an assistant will just as likely use them. But there may be some kind of SEO tricks that companies like that could use to make LLM agents prefer them.
That said, I don't know about most of you but booking a hotel / flight is the kind of thing I do maybe once a year - it's too incidental (for me) to even consider automating.
That said, Google (search) is in trouble, although they're ready for it because their own AI products are solid and they already have the infrastructure and whatnot to run them, unlike e.g. Anthropic or OpenAI for who their AI products are their whole business. Likewise, Q&A sites like Stack Overflow, Quora are at risk.
Even Wikipedia is seeing less visitors, but I'm not too worried about that because visitor count isn't what they need to (or should) optimize for - their main risk is articles getting poisoned by LLM generated edits.
Its probably just me, but I had to scroll all the way to the bottom, mentioning something about Macs, to verify that this was only about the Apple app store.
I’m an indie developer in one of the categories mentioned as being AI heavy (Photo & Video).
I’ve been shipping apps since way before vibe coding (https://heliographe.studio), and it’s been a pretty speedy transition in terms of how the market has shifted in the last 12 months.
Making a high quality camera/photo app used to require a lot of fairly advanced skills: UI design & engineering for the core app, an understanding of signal processing/color transforms/etc for producing the final image, graphics programming to keep everything rendering smooth in the preview (especially if you have advanced features like focus peaking, scopes/histograms, etc)…
It used to be that if you shipped a polished camera app, you would get at least a bit of attention just on those merits.
But in the last year it has become utterly trivial for anyone to make a camera app, and every week there are dozens of new entrants on the market (most of them claim to do their own unique non-AI, film-like processing).
But of those, which has gotten the most attention? I’d argue it’s the just released Chromagraph, designed in part by Rian Johnson of Star Wars fame.
If you compare it to every other camera app released recently, I’d argue it’s pretty average in terms of features/output quality. But that name recognition is what drives downloads more than ever.
Recognition and marketing will become key in the future if you want whatever content to get attention, maybe trusted reviews. When it become so easy to pump out whatever app, music, video etc. we will be drowning in it - and you need an already established named or a trusted reviewer or crowd-wisdom (as in app rating) to pick what is actually good.
Which finding is more surprising: that nearly 10% of charting apps mention AI, or that around 20% haven't been updated for at least a year? What does this tell us about the long-term sustainability of apps?
The 'abandoned' apps are also often actively advertising, so possibly they do something that causes it. I see tens of thousands of apps churn like that a month. They're not some quite app that a developer stopped updating, but something actively having ads bought for it that is then taken off market.
I wish there was a filter in the app store to filter and only show apps that having nothing to do with AI. It's become increasingly hard to find a decent, non-AI app these days.
Ah yes, I remember those times.
"bird anger"
"super pdf toolbox"
"mega vpn gold"
"speedy and very angry"
"six nights at Frankie's"
"minercfrafts"
. . .
45 comments
[ 2.4 ms ] story [ 26.4 ms ] threadCould bé another similar report about Google's App Store?
Thanks again!
It's wild to know that Freecash is somehow back in the App Store after [1] earlier this year and they're actively running aggressive marketing campaigns on TV again.
[1] https://techcrunch.com/2026/04/14/how-the-rewards-app-freeca...
I think most non-gaming apps will be replaced by a single AI app that is like a super assistant to you. No need to book on Booking.com app if you can just have your assistant look for what you actually want.
a) I won't trust booking's own assistant or "AI" and it will be garbage if/when they introduce it (despite fact that in order to handle it correcly it would need to have amount of information about me and my family and my trip that im not comfortable to provide them)
b) They will block 3rd party/independent assistant via CF or other measures - they want you to be on their property browsing on their terms upselling you features, not access via API/MCP
c) chat interface and general UI clunkiness related to AI and this kind of use case - when browsing booking I'm able to narrow to offers I want quickly, where accommodations are located on the map etc, lol, just thinking about doing all of that via chat UI and even more via voice interface makes me tired
Take all these basic lifestyle CRUD apps. Maybe it's a skill issue on my part, but whenever I've tried to use LLM apps over any sort of medium-to-long-term time scale for tracking something (ie. endurance running training), within about a month it begins hallucinating, recalling the wrong things, linking the wrong bits of data to other bits of data (dates, distances, heart rate data, workout paces etc), and it does all of this even when pointed to a basic spreadsheet or simple text file to keep it onboard with what happened where. It just seems to look at these files, and then look at it's own context and confuse all the details based on its own context.
Every other month, some brand new shiny model is released and it's exactly the same deal every time. I think the underlying technology simply can't deal with these sorts of goals. A whole range of apps for lifestyle tracking don't seem under threat, in my opinion, if this is the best its capable of.
Many other things I can't imagine wanting an AI to do for me, especially anything involving payments, bookings. There was a discussion the other day on here about AI being used to order takeout. Why would somebody want to delegate one of their last personal decisions to a LLM?
And there, it has a database of sorts, running god knows where on god knows what, and you have a single-user web app that you may or may not ever open and see (since your favorite AI agent can just read and update it for you anyway).
Today it’s still something the user has to suggest / request.
Tomorrow? If they put a bit of effort these models will reach for building these one-off ephemeral or long-lived hyper-custom apps the same way they now reach for a python sandbox for basic arithmetics and statistics.
Looking for a flat? Scheduled task + CRM like artefact for it to monitor, maintain, and update.
Looking for holidays? Your own personal shareable moodboard with destinations and what not.
Have a use case or task for which chat or editing text docs is a frustrating way of correcting or giving feedback? Have it build a dedicated, custom artefact just for that.
When you encounter a use case where LLMs limits show, make them build tools.
I wonder how open these services will be to data storage if they do become a "universal app". If I used LLM to make an app for tracking training for endurance running, and I slapped an 800kb .fit file in there each day, the storage for all of these files would probably be fine, but without condensing each run down to a textual summary like Garmin/Strava currently does ("You ran 10km in 35:00, at a pace of 3:30/km. This improved your speed/{vague motivational LLM quote}"), wouldn't the context become enormous after a pretty short duration, causing huge costs? Genuinely curious.
Harnesses have been developed for this, and its also fairly trivial to do large scale verifiable-rewards post-training for "personal assistant" style things. eg., invent a booking website, invent a PA GUI-aware harness, and reward the model when a successful booking takes place. Do that 1bn times.
However, about 1 year ago, AI coding agents were laughed. Now the vast majority of code is written by them.
Intelligence getting exponentially cheaper per token. At some point, there will be an inflection point where "it just works".
I may be completely out of the loop here, so forgive me if I'm missing details, but I always have and continue to believe that the coding prowess of LLM agents more-or-less comes down to the sheer abundance of data it has trained off of on the web, and that code itself is ideal for LLM by the way it is discrete and syntax-bound, same as the way LLM can write very well provided they've read more-or-less the same sequence of text before in abundance. For the speed advantages it provides in coding, and for the accuracy it has for well-established information, LLM is a very useful technology.
I just don't follow how they are going to do the same thing with such granular telemetry data, such as datasets from a day of nutrition or a workout. The values are continuous, minute, and are so parallel to noise in the way they are provided. Take running. If you dumped a .fit file from a Garmin or other watch with a series of heart rate values, paces, elevation levels, GPS co-ordinates, respiration rate, vertical oscillation etc, and you trained on a bunch of files like this, the values are so discrete that I can't understand how it would connect them and tokenize them across different people or even the same person's various workouts, with the high degree of factors (nutrition, climate, illness/injury state and severity) that can alter those values. It's no longer about just tokenizing text, but with advanced calculations.
I've seen second-hand through friends in running with apps like Runna that it produces plans that aren't suited for them at all. There's simply too many small variables that these things don't seem to be able to map. It gets caught up on summaries of runs. It can give general advice about things, but it doesn't have the level of detail or understanding necessary to give good guidance. Strava "AI" and Garmin "AI" just make a 2-3 line summary from the average data from the entire run or workout.
Perhaps this is something that is already being worked on, and I would be happy if it was, but I just can't personally see how LLM can work with such datasets in a detailed way, beyond the useless summaries Garmin and Strava give "You ran 10km in 35:00, with a pace of 3:30/km. This shows that your speed is improving". I would love an explanation/links on how LLM can get around this if anybody is willing to spend the time providing it. Happy to have my mind changed.
For example, I use LLMs to create 3d models. I don't tell them to recreate the 3d vertexes every time, they have a python script that programmatically creates the shapes, and with a freecad mcp, they get views in images, also they can create other scripts to check for collisions and other issues.
They can also get a histogram, discard noisy values, check outliers, etc, etc, using scripts or by running a traditional ML value to get later the refined data.
Opus 5.5 is quite decent currently, but expensive and not enough for many challenges, but I see the progress happening.
The problem with these Garmin and Strava AI summaries is that most of them just point an LLM at a heap of data and ask them to offer a summary. They also usually use older models since the feature is not something particularly well curated. I do financial modeling with LLMs and use the LLMs to build/verify/iterate on models.
If you're curious on what it takes to run long-running tasks Anthropic has https://www.anthropic.com/research/harness-design-long-runni... out.
Browsing through hotels is one of the best parts of the entire experience.
I prefer the part where you wait in airport or on train station over it.
There never was any need for these apps. A website works perfectly fine for simple use cases like that.
That said, I don't know about most of you but booking a hotel / flight is the kind of thing I do maybe once a year - it's too incidental (for me) to even consider automating.
That said, Google (search) is in trouble, although they're ready for it because their own AI products are solid and they already have the infrastructure and whatnot to run them, unlike e.g. Anthropic or OpenAI for who their AI products are their whole business. Likewise, Q&A sites like Stack Overflow, Quora are at risk.
Even Wikipedia is seeing less visitors, but I'm not too worried about that because visitor count isn't what they need to (or should) optimize for - their main risk is articles getting poisoned by LLM generated edits.
<https://obelisk.club>
I’ve been shipping apps since way before vibe coding (https://heliographe.studio), and it’s been a pretty speedy transition in terms of how the market has shifted in the last 12 months.
Making a high quality camera/photo app used to require a lot of fairly advanced skills: UI design & engineering for the core app, an understanding of signal processing/color transforms/etc for producing the final image, graphics programming to keep everything rendering smooth in the preview (especially if you have advanced features like focus peaking, scopes/histograms, etc)…
It used to be that if you shipped a polished camera app, you would get at least a bit of attention just on those merits.
But in the last year it has become utterly trivial for anyone to make a camera app, and every week there are dozens of new entrants on the market (most of them claim to do their own unique non-AI, film-like processing).
But of those, which has gotten the most attention? I’d argue it’s the just released Chromagraph, designed in part by Rian Johnson of Star Wars fame.
https://www.theverge.com/tech/1025659/rian-johnson-chromagra...
If you compare it to every other camera app released recently, I’d argue it’s pretty average in terms of features/output quality. But that name recognition is what drives downloads more than ever.
For example these apps advertising in October 2026 https://appgoblin.info/reports/ad-user-acquisition-2026-octo...
And some of them, like the block game are already gone:
https://appgoblin.info/apps/com.game.best.mini.pro.dragon.ra...
This link is currently not active: https://play.google.com/store/apps/details?id=com.game.best....
What could be those apps:
- resale/second hand apps that offer boosts to be seen?
- apps that sell alert systems when a price is dropping?
- grocery/listing apps that sell "no ads" functionalities?
- gated brands with VIP things?
Yet, 11.5% looks enormous. I may be old.
It is not as if slop was AI invention...