I suspect AI is going to result in the bottom falling out of the market for simple apps.
React made simple webapps a case of just gluing dependencies together and much more approachable to a generalist developer than the previous generation of web development was. But native apps weren’t affected in the same way. With AI I suspect we’ll see a lot of simple apps, the ones that really aren’t doing much other than CRUD operations on a remote API, become very heavily AI generated by generalist developers.
But there will still remain a healthy market for working on considerably more complex apps.
Will the complex apps be discoverable though, and if not, will the devs bother at all? I feel like this would be a win for old-school open source, where people would do it for the love of the aim, and better vetted app stores like F-Droid might win out
Surprisingly I'm seeing a rapid increase of discoverability for my apps FROM LLMs. The flood of Claude/ChatGPT users finding my apps right after I get a basic site up, before I've told anyone it exists, is astonishing.
If you are solving a problem users are searching for, you will be found and found faster than ever via LLMs.
You're overestimating the extent to which individual developers have a choice here. My employer signed up for a Claude Code membership, I use Claude Code. I cannot use Codex.
Anecdotally I hear of folks with workplace Claude Code subscriptions all the time. I'm not sure I've ever heard someone talk about their workplace Codex subscription. Anthropic clearly did a far better job chasing corporate customers while OpenAI was busy chasing consumers with Sora etc.
The OP seems unaware that Claude had a lead in this space and captured market share and attention for that reason alone.
The test they (supposedly) ran with their coworkers to look at PRs from both is such a bad way to compare LLMs that I don’t think they’re very experienced with using them.
Having a lead in a new market even for a short time creates initial conditions that are in your favor. That’s why startups get all fired up over being first to market.
I know it is irrational vibes, but how the whole Department of Defense situation was handled will always make me partial to Anthropic. Unless there so huge shift, of course.
No, it wasn't and no one who used Codex and Claude since release could seriously make this claim. It was a magnitude behind Claude and only catched up recently
> The OP seems unaware that Claude had a lead in this space
I remember using GitHub Copilot (OpenAI "Codex" mk1) in Aug 2021 (ChatGPT would launch a year later 2 weeks after Meta's botched Galactica release). Cursor & others took it and ran a mighty good race.
Okay so regardless of the model, platforms provide attestation from end to end that nothing is being logged from either input or output, including at the firmware and OS level to the extent that the customers have proof of the data never being saved. AFAIK, both GPUs and TPUs support this.
Intellectual property. My employer has an agreement that our code will never end up as part of Claude's training data. At this point there are also now custom Claude integrations etc.
I'm sure they could also negotiate a similar deal with OpenAI but in my outsider experience it seems that negotiations around these kind of corporate contracts takes forever and when the selling point is "they're broadly pretty similar" I suspect the motivation isn't there.
> My employer has an agreement that our code will never end up as part of Claude's training data.
Bit of a tangent but it is funny to me that so many companies talk about how some large percentage (sometimes 100%) of their new code is LLM-written and they still bother to worry about their code being used for training.
If an LLM is writing all your new code then your existing code certainly wasn't some unique special secret, and your new code came from the LLM to begin with.
Claude Code was a huge huge huge step up when it came out, absolutely massive.
It was barely marketed. I always turned copilot off, never found any benefit from Cursor. Claude Code was vastly different in conception, function, and capability, a product that defined an entirely new category of product.
Perhaps to others, that found copilot or cursor useful, it was merely marketing. But to me it was function and productivity, that I had never seen before.
People try to dismiss these things as LLM wrappers, but the LLM will be commoditized, and the wrapper will be where the real product design goes and where the real differentiation happens. Owning that unique process of communication between the dev and what the dev wants, figuring out the most stuff with the least complete spec, and maximizing every bit of the very tiny communication channel between the dev and the LLM and the code on disk, that's where 2026 and 2027 will be focused, until the next category defining product is created.
I started using Claude web with sonnet over chatgpt before any of the coding tools came out and noticed other founders were using it too and the reason was pretty simple - it was much less likely to hallucinate non existing APIs than ChatGPT
I have lots of choice (I own the company) but I'm still not going to switch from Claude until I see evidence that the alternative is meaningfully better. So far I don't see that evidence. In the past I've looked at using competitive products and it turned out to be a painful experience (Cursor didn't work at all on my computer, Google thing -- whatever it was at that time -- required dependencies I wasn't willing to install). I'm sure these issues have been resolved since but why would I spent time kicking the tires of another product just to have it work "as well"? Claude's cost to me is minimal so there's no cost savings to be made.
fwiw nobody "marketed to me". I picked Claude because friends were using it with great success and they helped me get started with suggestions on prompt style. Before that I'd played around with various LLMs for coding but not done any actual production work.
Fascinating. I've written cross platform (WASM, iOS, Android) libraries with Rust before and had a good time but Rust can be a pain too. Cross-platform Typescript is a really interesting proposition.
That said, the more I think about it the more dubious I am. The site boasts no runtime dependencies but clearly it’s going to need things like a garbage collector, you can’t just magic that requirement away. At a certain point is it just doing what a JS engine’s JIT compilation does… except ahead of time?
Also doesn't inspire confidence that the text on the site is very clearly AI generated and the GitHub log shows an endless stream of AI powered commits. About 15 per hour, every hour? Doesn’t scream stability.
Less of an accident than a byproduct of the unique ecosystem the web lives in (which is a positive!) Compare to say, backend development. If I said database indexes are deeply inconvenient and that I shouldn't have to make them I'd get laughed out of the room and justifiably so. But by comparison when a developer says "I don't care to learn CSS" folks nod their heads in agreement.
Correlation != causation. There are a ton of differences between the US car industry and those in other countries, unionization is just one factor.
As a counter anecdote I’d point to Boeing’s non-union facilities, which have produced notably less reliable airplanes than their union locations ever did.
Boeing actually offers a fascinating direct comparison. 787 Max has historically been assembled in two locations: unionized Renton, WA and non-union Charleston, SC. The Charleston planes were notorious for needing rework at Renton before they were airworthy.
But the conclusion is muddied by the fact that Boeing has been making planes in the Seattle area for a century, so the talent pool is larger and more qualified than those they could find in or persuade to move to Charleston. Also, the whole Charleston move was one of many drastic cost-cutting efforts, including the spinoff of Spirit Aerospace that ultimately led to the door blowout on the Alaska Air MAX 9.
Eh, I think the framing isn't quite right here. The Neo is a wonderful machine but if you want to upgrade it you're out of luck, the damn thing is sealed shut. By comparison the Framework lets you upgrade individual components over time to keep your system up to date without buying a whole new one.
Maybe that doesn't matter for the godson. But it's an important differentiator: the Framework is a (semi) premium product with premium features. If you don't intend to use those features, paying the premium rarely makes sense.
I think this model works for the 13 and 16, because you're already buying a good laptop that you can keep longer by upgrading. The 12's base specs and more than that the experience is pretty bad. The screen and speakers are terrible.
The 13 also targets people buying it for themselves and who value ownership. The 12 targets the education market and how many 14 year olds are sensitive to ownership, repairability and e-waste? If they are they would probably get something better second hand. You'd have to have a parent that is sensitive to this issue and is also willing to force down this bad laptop onto their children instead of whatever they prefer.
I love Framework, and the bet to try to win over the education market was worth making but the execution is so poor that I don't think it works out.
The MacBook Neo will happily last you the 4 years of highschool and maybe your bachelor.
The 12 for me has a very strong appeal as a smartphone / tablet replacement.
I've had smartphones and/or tablets for approaching 20 years now, and they've always struck me as very frustrating compromises. Mostly Android, but some use of iOS as well, and yes, the OS (in both cases) is fundamental to the limitations.
I've also used MacOS heavily (I'm on it now), and I don't like it, relative to Linux.
The Framework Laptop 12 is smaller than my most recent tablet (a 13.3" e-ink), though somewhat more massive. It frees myself from a plethora of Android limitations, crapware, inconsistencies, and the non-repairability of the hardware itself (presently an issue). It gives a real-computer experience, with some compromises for size, but I'm pretty sure that's a net win.
Paired with a limited-feature phone and possibly a few dedicated devices for specific uses (camera, audio recorder), I'm good.
And the 12 should provide an easy decade of service.
> The Neo is a wonderful machine but if you want to upgrade it you're out of luck, the damn thing is sealed shut. By comparison the Framework lets you upgrade individual components over time to keep your system up to date
The Framework 12 in the story costs $799, a $300 premium over the $499 MacBook Neo.
So you're paying an extra $300 up front for the option of spending more to upgrade it in the future, and getting a slower computer during that time.
That's a 60% premium to have the ability to upgrade a slower laptop.
Alternatively, they could sell the MacBook Neo for $200 in a couple years and buy a next-gen MacBook Neo and they'd still come out ahead.
Some people value upgradeability to an extreme, but I can't see a justification for spending a 60% premium to buy a worse product just to be able to maybe upgrade it in a few years. This is a starter laptop.
That might be true to some extent but what about the current product? It's nice to tell yourself that you can upgrade it in the future but the best of what the product is today isn't a great value, will the future upgrade make it better? Should we purchase a product today on what it might be tomorrow?
I think Jeff is correct when he says, "for an overall worse experience, are you willing to pay 20-40% more?". That's a tough sell. I think the only reason for me to take the Framework 12 over the Neo would be because I want to advocate for a world where upgradability and repairability are common things.
I don't think the idea is that the upgrade will take it from decent to stellar compared to other things you might be able to buy for the same money, it's about paying a bit extra now to be able to go from decent-in-2026 to decent-in-2031 while paying a fraction of the cost that you would buying a full replacement in 2031, not to mention saving a bunch of waste. And then in 2036, and 2041, and 2046... They haven't been around long enough to be confident it'll work out that way, but that's the bet in my mind.
For lots of things, yeah. Don't try to fold proteins or open Facebook or whatever, but if you want to run a 3d printer or make some drawings or organize a bunch of notes and docs it would be way more than enough.
The neo isn't upgradeable, but it also isn't sealed shut. It's actually one of Apple's most repairable devices. If I were in the market for this class of device, I personally would still go with Framework for a variety of reasons, but I still think it's important to give apple praise for the pro-consumer choices they made (and probably could have gotten away without) in the Neo.
While I agree with what you’re saying the typical AI agent doesn’t say “I’m not totally sure about this, should I search the web?”. It often just spits out a reply based on its knowledge.
That was true a year ago, I don't think it's true today. I can't remember the last time I saw Claude or ChatGPT confidently answer a question that they should have searched for instead.
If you watch their reasoning traces they often say things like "this is a well-known historical fact so I don't need to search for it", or more frequently they spit off a bunch of searches.
I had an issue with Claude Sonnet the other day in one specific chat where it made shit up without searching, I called it out, and then it searched, got the answer half right, and made up some more shit. It kept doing this several times in the same chat. Didn't have any issues with any other chats, though.
Anecdotally, it still happens a ton to me. They also still make super simple logic errors that they immediately reverse when pressed. For example, I asked Opus 4.7 last night how to cool off my room without making it too humid inside (indoor temp 78°F, humidity 45%; outdoor temp 64°F, humidity 99%). It suggested opening a window and assured me that the humidity would not rise above around 60% which would still be comfortable. I asked it to justify that and it said:
>You're absolutely right about the humidity — I was sloppy with that aside. If you ventilate enough to meaningfully cool the room, you're replacing indoor air with outdoor air wholesale, and you'd converge on outdoor conditions: 64°F and near-100% RH. That's miserable. The 55-60% figure I tossed out was hand-wavy nonsense — it would only hold if you barely cracked the window and mixed a tiny fraction of outdoor air in. At any ventilation rate that actually cools, you're just moving outside air inside.
I don’t think those are really comparable. The blockchain was trendy hype, relatively few companies actually adopted it. Where did Netflix use the blockchain? Google?
By comparison almost all tech companies I know have leaned heavily into AI.
React made simple webapps a case of just gluing dependencies together and much more approachable to a generalist developer than the previous generation of web development was. But native apps weren’t affected in the same way. With AI I suspect we’ll see a lot of simple apps, the ones that really aren’t doing much other than CRUD operations on a remote API, become very heavily AI generated by generalist developers.
But there will still remain a healthy market for working on considerably more complex apps.
reply