At the end of the day, no matter how you put it, someone has to still actively ask for those solutions. When we get to the point that the saw decides to activate itself and go cutting stuff without being asked, then we can say the tool is actually doing stuff.
It's not really an apples-to-apples: you can ask an "AI" to build software to do a particular thing, and it will happily oblige and spend some of your money on it and produce something that actually runs.
You cannot just turn the saw on (if electric), point it at some wood and expect anything resembling furniture.
Comparing AI to a saw it's like admitting it's the same as an IDE. To some extent it is, it is more autonomous too though. It's more like a brain child robot with a saw and a grin on the face
Comparing AI to simple tools like a hammer or saw just isn’t right.
AI tools are complex systems that definitely can have agency, even though that is mostly human controlled. Bad controls = bad outcomes.
Yes, I totally agree that humans have the most control over them, but that doesn’t mean it’s completely off the hook.
If we really have to draw simplified real life comparisons, maybe a car or a gun would be more suitable “systems” that we can metaphorically compare AI to.
The plain text editor never produced any kind of logic, or any kind of text for that matter, beyond what was explicitly typed into it. We never said the IDE "does" X when it got auto-complete.
There was a time when people had to walk around in a room flipping switches to create routines that were combined to produce software. Then there was creating and feeding punch cards or paper tape into machines. Go forward some more, and there was typing into a writer (couldn't even see what was being written!), and running through an assembler. And then there was the text editor, into which one typed (finally can actually see what's written so changes can be made) and saved, then switched to a terminal and ran a compiler (and maybe linker) on the thing saved. After that came the IDE which was not just the text editor with a "compile" menu option, but there was also a lot of templating and auto-complete and external libraries. Today there are coding agents bringing all this together even more with a few instructions.
Know what changes? The amount of human effort needed to produce increasingly sophisticated software keeps falling. And know what doesn't change? People are still inputting requirements into the machine, testing what comes out, and iterating until it works as they'd like it to. That's the software production process in a nutshell. Doesn't matter if it takes a few thousand iterations over many months, or a single iteration that takes a few hours. It's still people who actually produce software, not the IDE nor the LLM (both of course generate code to help in the production process).
> And know what doesn't change? People are still inputting requirements into the machine
Before LLMs, only incredibly few programs were written by "inputting requirements into the machine", most code was about inputting instructions that fulfill requirements. Usually code is/was related to the requirements, but for anything non-trivial it was almost never a direct translation of the requirements.
So as far as I can see, you're trying to gaslight here.
Given the context, there really isn't much difference between "requirements" and "instructions", per se. It's the result that matters. Flipping switches, using punch cards/tape, typing hex/alphanumeric characters or using speech with coding tools all result in code.
Few programs were written in the past because the effort required was huge. Can you imagine how long it'd take to correctly code a simple calculator using only 0's and 1's? Or even hexadecimal? And then suddenly it took just a few lines of C. The transformation scale is similar now we can use natural language instead of having to learn a programming language.
You're pretending there's not much difference between "requirements" and "instructions", but there very obviously is - it's the entire goddamn reason programmers were/are needed. If you were correct, there'd be no reason to spend so much money on people who translate requirements into instructions, but that's not how the absolute majority of our industry worked before LLMs.
> Few programs were written in the past because the effort required was huge.
What are you talking about? A lot of programs were written in the past.
I can't tell if you're being deliberately thick or are just incredibly good at missing the point. I'd think it's pretty obvious that there's little distinction in my use of "instructions" vs "requirements": whatever it is that you give a machine, which it then assembles/compiles/interprets in such a way that you can use the output for some task (really can't think of any better way to break this down). There's little distinction because that thing you give, whether it's in machine, a programming or a natural language, yields the same result per intention like "good morning", "bonjour" and "ohayo" all elicit a greeting result.
The primary reason why programmers are/were needed is because humans and machines speak different languages natively, and someone/something needs to bridge the communication gap. That bridge is done via translating instructions/requirements into the many programming languages we've developed. Now we have LLMs to also do the translation.
> A lot of programs were written in the past
Yes, comprehensively, but also only "recently" (given any 2+ slices of advancement in tools used for software development). Somehow you missed the - I think - very clear "0's and 1's"/"hexadecimal" -> "C" and "programming language" -> "natural language" references.
> I can't tell if you're being deliberately thick or are just incredibly good at missing the point. I'd think it's pretty obvious that there's little distinction in my use of "instructions" vs "requirements": whatever it is that you give a machine, which it then assembles/compiles/interprets in such a way that you can use the output for some task[...]
I understand you perfectly well, I'm just calling out your misrepresentation of these words, because they are very much distinct in our industry, even today. Trying to say differently is simply gaslighting.
It isn't a misrepresentation. At every point it was human communicating with machine. That's the context set from the very beginning. Whatever the language used in that communication, whatever the actual input provided is called, however many layers of translation between the input and the output, it's all the same result: things that map user input to machine output. And to the original point, humans are integral to the production of said things.
If programming isn't communication, then what is it?
"Communication is commonly defined as the transmission of information. [...] A common way to classify communication is by whether information is exchanged between humans, members of other species, or non-living entities such as computers." - https://en.wikipedia.org/wiki/Communication
Notice how it doesn't talk about programming being a "composition of sequences of requirements"? Or how it's not defined "as the transmission of information"?
None of the issues raised in the post/article here even require any particular knowledge of how to build software. Anybody with a basic attention to detail would have noticed them. It's just somebody who simply DGAF shipping to live.
Nuclear weapons are a totally neutral technology with no particular affordances. That they can be used to kill people is totally beside the point and irrelevant to any discussion about their merits and downsides.
In particular they offer their owner a high maintenance decaying shelf life, or an explosion.
Of the 2,476 explosions to date, two were used specifically to kill people, some of the others did kill people, although such things were supposedly unintentional.
They have not afforded a better way to make canals, mine in a controlled manner, or fart a starship to space .. although people have desired such affordances.
I just used Claude opus 5 high to refactor our permissions model to an hierarchical one. In the validation function, if a part of a permission grant was missing in the hierarchy, claude happily granted them approval instead of rejecting, fail open for permissions.
Always review the code an LLM creates.
(When I told Claude that this is very bad, its reply was “Good catch!”)
This perspective on vibe coding is both funny and contradictory. The core issue with vibe coding is that people ship things they don’t understand. To flip it around and blame it on the humans involved only after something goes wrong as if the issue is them vibe-coding “wrong” is foolish. The issue is the vibe coding.
For the vast majority of these issues, certainly including those mentioned here, simply testing before going live would make them beyond clear. So it's simply a bad developer likely working under bad management and, shockingly enough, producing a bad outcome.
LLMs will be an interesting test of the current broken systems at most large companies. You can literally fire your entire engineering team and replace them with your "really smart" cousin and Claude. And that's going to be mighty tempting when the MBA mantra seems to be that cost reductions this quarter are infinitely more important than any potential costs in the future. But of course they're not.
So as I understand it, vibecoding means (meant?) that you don't look at the code it produces, you case only about the end results, e.g. website through a browser.
But in this case, they seem to have skipped that as well.
Vibe coding is just a really big hammer, what you do with it and what you do with results is still up to humans. Even if this was result of it, fuckup is ultimately fully in hands of somebody responsible (for product, website, clients etc).
If not, company is not viable and will crash badly soon. Market forces and all that.
If I had a dollar for every RoR coder who didn't understand the internals of the interpreter I would still not be richer than Musk because there are not that many coders.
No one at any company has ever understood every bit of the stack they ship; see all the bugs AI is finding in all the frameworks and libs we been shipping this whole time; oh yeah we sure knew what we were shipping.
git pull git@github.com:allofit/allthethings.git engineering patterns is what created this mess to begin with
Software is hard. But especially, evolving software over many years is hard.
I've seen people build great software but fail to evolve it.
All the lessons from agile, clean code, TDD and testable architectures, CI/CD, DevOps are really about building software that can evolve, yet most of those get ignored from time to time due to cost (frequently called time) pressures.
GenAI stresses this loop even more, and no human can effectively deal with the level of output you need to fit in to continue effectively shipping. This can only mean more bugs rather than fewer bugs.
Yes, in a well executed project, LLM agents can bring immediate value and find and fix bugs: because the system has the good feedback loop built in. As soon as you lose that (over shorter time with LLMs than with humans), it starts getting exponentially worse.
And this is all even before you stop and think if you are even building the right thing.
I have noticed that the code harness is extremely willing to run me into extreme levels of decisionmaking detail if I ask it to defer substantive decisions to me. And then, for each answer, it’s happy to dispatch (token-gobbling, but often insightful) subagents to study the repercussions and argue back with me. To nitpick intersections between my decisions, my stated core principles, and the standing rules for the project. And to point me at standards to fill in the gaps in my understanding.
It’s exhausting, but so often, my mush-brain has to concede that it has a point.
On any given day, several hundred decisions in to what I’d imagined would be a sensitive but straightforward spec, I’m keenly tempted to just accept its default judgements…
By now, I’ve encountered the whole of all 3 official bodies’ standards in anger. For sure that’s not the same as real understanding or hard-earned experience. But I sure understand the formalities of the problem space a hell of a lot better than I did when I started this particular vibe coding adventure.
Contrast with my young friend whose approach was “make me a first person shooter, in Los Angeles, I want {some kind of weapon} and cars, make it super polished and fun”. And his follow-ups along the lines of “I don’t care, make it awesome, whatever’s most awesome.” I hate to say he’s holding it wrong (as he did get some sort of Unity/Godot thing out of it), but… does my neuroticism elevate my LLM-hectored coding out of “vibe” territory?
I'm sure it's both different groups of morons, but the constant shifting back and forth between "if you're reading code in 2026 you're wasting time" and "you wouldn't blame a hammer if you hit your thumb" is absolutely hilarious. LLMs are either a simple tool or a hyper-intelligent system depending on which characterization is anticipated to sell the most snake oil in any given instance.
At the very least, if these are indeed different groups of morons, one group is shutting up whenever the other is spewing their flavor of bullshit.
Was going to say the same. In fact we're seeing a huge wave of hacks caused by old human coded systems getting owned by "vibe" attackers.
Vibe coding is such an overused term now. Stats say what, 80% of programmers use AI to generate code? 90%? So everything is vibe coded. If you do well, no, it's still engineering. Someone makes a mistake? Cuts a corner somewhere? Vibe coded.
Because it allows you to produce so much code that it's not viable to carefully review each line/module, and get the benefits of the fast output.
Writing code yourself forces your brain to start creating pathways in your brain and you can verify the correctness of those pathways in your mind (e.g. am I actually doing this right, if I have seen such API would I be happy to use it,...)
Not saying we should write code manually, just showing the differences.
And even if you don't carefully review each line of code (or if you do), it helps a lot if you understand the basic principles of what you're trying to build. In this case, the vibe coder clearly had no idea about account security standards.
Programmers don't just write code. They also understand why they're writing the code and what it's supposed to do.
Because if a human was to program it, they might have stopped for a second and thought "Is this really a good idea?" whereas the LLM just goes "You're absolutely right! Here's some updates to the security code:"
I don't think it does, I doubt SOTA models are going to sign off on this.
On the other hand, people incompetent enough to ship this couldn't have produced a working website if they had to handwrite the code, so in that sense LLMs are at fault.
My SOTA LLM spends disturbingly much time stopping me and telling me to shape up and do things right, especially on the security/legal/privacy side. Speaking from recent experience, it would NOT have anything CLOSE to this—I’d expect it to literally refuse to ship.
Whether that says more about me or about LLMs These Days, I’m not sure…
Wow, that's an impressive downgrade. Not only is the site uglier (gradients, gradients everywhere!), but I could log in without needing my 2FA at all. Shame too, since I found the old Humble site pretty charming.
I expect marketing types are freaking out even more about the jobpocalypse than devs. While it takes some skill to determine whether resultant code is working, marketing goes by vibes all the way up the chain.
I've been involved in many "rewrites". Almost exclusively, the result is worse for users at release.
Especially with "big bang releases". Where a team is working behind the scenes for months or longer, attempting to rewrite the old system "bug for bug". Releasing and then finding out that some users would "hold the space bar to make the CPU heat up, and configuring emacs to interpret that as 'control'" (XKCD 1172). First, users don't see any improvements for all the time that the new system was being built. Secondly, they get all the new stuff bundled. So while they may even get value from the rewrite, it is bundled with the stuff that makes it worse. Three new features, one new bug. And all everyone sees is that new bug.
But even rewrites where parts are swapped out (strangler vine pattern!) are worse for the users. They do get value fast. And in small increments. They can give feedback early. But the (much longer) period in which the replacements are implemented, is almost always one of confusion, complexity, and "having to learn new things every week". We all know the systems where "change email" is a link that leads to a form on some other site, in an old design rather than part of the "edit account" page. Because the "account service" has been extracted, but the "notification service" replacement on the backlog still.
"rewrites" are almost always for the benefit of the business, the technical team, or other internal reasons. If "benefit to the users" is a part of it, it's almost always secondary: so that we can build and release and scale new features and cool stuff for users faster than with the old legacy spaghetti debt.
As if anyone holds management accountable either. Big companies are full of VPs and Directors who make a mess and transfer orgs just before it comes back to bite them.
I don't know about illegalness, I pressed "Get your games" button accidentally and it immediately deducted 12,99€ from my Paypal account without confirmation. What the hell.
I recently wanted to download a game I bought a long time ago through them (Machinarium, new macOS builds). Unfortunately the file on their CDN is corrupted. Their reply, after a week's worth of back and forth: sucks for you, it was sold too long ago so we won't bother fixing the issue.
The article doesn’t seem to provide any direct proof the new site was vibecoded? I mean, probably it was, but people were writing insecure software long before AI came along.
Without evidence, accusations like this are, ironically, just vibes-based.
What other reason would there be for a complete rewrite without any sort of security? They had a fully functional and secure backend, why would it require replacing just to redesign the site?
It was ok before, not perfect but more than ok and secure. The new version it is indeed vibecoded doesn't matter if with ai or not, vibecoding to me means careless and sloppiness, either it be human or not
You get regressions in design and security when you:
- never documented how the old site work
- never asked a good model/person to comb through the old codebase, and write a spec + PRD for how it works
- ask a cheap LLM to redesign the whole website (without migrating auth, or some material % of old content)
I wholly agree with this. People already took shortcuts with passwords, excessive MFA causes the same _need_ from people to flow through it, and we see it everywhere already (addons to autofill MFA etc).
> And yes blizzard my wow account is not worth anything either.
Plenty of other players would strongly disagree. Blizzard pushes so strongly for 2FA because the support costs associated with people complaining about their WoW accounts getting compromised and drained are massive. If every player were to just shrug and start over from scratch, Blizzard wouldn't have bothered with it.
For a lot of services I'd rather just get a 'magic' link in my email and not have to deal with a password at all. I've got good email security so it's worth hanging a few lesser services off of that.
> And yes blizzard my wow account is not worth anything either.
You’d be surprised; a friend of mine had a battle.net account under one of his emails, because apparently his email was used to make a WoW account in college and was banned. He had never played wow before, contacted support, and they didn’t say what happened but revoked the ban
I'm not going to blame the tool. I'm sure as fuck going to blame the peddlers of the tools that are always ready to support this kind of nonsense; worse, they tell everyone around that this is perfectly fine.
I'm not blaming the bulldozer. I blame the people who gave everyone bulldozers knowing that they have faulty steering and that nobody is used to bulldozing. And the people who shut down every attempt to prevent that with the argument "bulldozers are the future, you can't stop them!".
And yes, I also blame the people getting on that bulldozer, bulldozing like there's no tomorrow. But mainly I blame the first two groups.
Bulldozers aren't restricted vehicles, anyone can buy them. What you seem to be upset at is that they became very cheap and easy to obtain, and everyone is using them to landscape their small front yards and the damage caused is often outweighing the time saved by using something simpler that takes more time and effort. The bulldozers work fine, but if you use a large one to clear everything you might just find you've also taken out the walkway to your door inadvertently, when a small one that took a bit more time and effort would have allowed for you to target what needed attention.
That metaphor has been stretched about as far as I can take it, but it's the tools and people getting access to tools they've not been trained on to do tasks they have no prior experience with. As someone who just laid laminate flooring myself for the first time, I'm keenly aware of both having the right tool for the job and the knowledge of how to use it and knowledge of how to do the task in general affects the end product. The mistakes I made are my own, not the cause of the incorrect tools I used at times.
Once you can't go for a walk without experiencing a bulldozer bulldozing the wrong thing somewhere on your path, it becomes a distinction without difference.
The bulldozer goes in the direction the person is sitting enough of the time we can call it "all the time" without having to argue "what if it breaks down, or a lever breaks and sticks on".
> I'm more inclined to blame the usage rather than the tool here.
Which is what your parent comment is doing. They didn’t say AI was to blame, they said this wouldn’t have happened without it. It’s likely your fault if you blow up your hands with a firecracker, but it’s also true that it wouldn’t have happened if firecrackers weren’t a thing.
> Do we blame the bulldozer if it bulldozes the wrong thing?
That’s such a tired analogy. By this point it’s honestly hard to believe it is made in good faith.
There isn’t a massive industry with tendrils in every facet of society trying to sell you on autonomous bulldozers that do the bulldozing for you without interaction, is there? Caterpillar is working on autonomous bulldozers, but even those aren’t targeted and advertised at every household as something you have to use.
Disagree on which end of the telescope were looking down.
AI has caused a generational set back on how stakeholders view the value of expertise knowledge, but the gun is still people, AI is just one of the bullets along with things like obsessive cost cutting.
My large orgs entire marketing strategy is now run by an MBA using Claude to generate 'idea and best practice docs'.
The process that lead us to not having a CMO or any other marketing leadership with domain knowledge? Consulting partnerships using MBAs.
The reason we're not changing any strategies even though we're burning money pper CPA? MBAs concerned seeking alignment and having the right narrative to change course.
That is true, but this can be said about so many things. Hit and runs would also not exist if it weren't for cars.
The technology itself is not to blame, but maybe the marketing around it should be. AI coding tools are marketed as magic tools that can somehow spit out perfect solutions even if the input spec is garbage. Of course that is impossible, and no amount of technology can solve that problem unless every AI coding agent is equipped with guardrails that makes it stop if it gets the impression that its user is an idiot.
Thought experiment. A startup is at the end of their runway. Need to show income or go down. Businessmodel is to sell "one off" sales, and vast majority of their accounts buy once or twice (a year or ever).
What if they "accidentally" make it easier for hackers to take over those accounts. Accidentally, as part of a large overhaul or rewrite of their systems.
Then, at least for a while, their accounts suddenly "buy" much more products per month or year.
Can someone there please just click the rollback button now and we can all forget this ever happened. I don't think that new site can be saved. The old site was fine. Just leave it like it was. I bought a bundle as recently as last month. Three bundles I think.
If they just focus on finding more fun bundles to sell and don't touch the site code again, ever, it will be fine. I have been happily buying things I usually do not need from them since Humble Indie Bundle 1.
This is something I don't get. The old development process was slow, but it was mostly safe and reliable.
People are rebuilding working things from scratch with AI just because they can.
AI eroded the greenfield advantage of rewriting a project from scratch because a rewrite generally meant that the code went through the fingers of the new developer. With AI that advantage is gone so you are better off incrementally improving the old code base to minimize overall maintenance costs and reduce the needed headcount as a result of this work.
After they repeatedly reinstated, without consent, recurring payment for their subscription plan Humble Choice, which I repeatedly chose to pause (yes, they offered it as a first-party option), I had to cancel the plan entirely, as an early subscriber back when it's called Monthly.
It showed me that they just don't care anymore.
But it still baffles me as to how an LLM could make it this worse, when they already had decent security measures. No top-of-the-line model should've done this. I guess it can only be concluded that they didn't even care to spend some money on SOTA models, even DeepSeek 4.1 Flash, instead turning to the cheaper but also much less capable models to save a few pennies.
- The UI components are the ones that cheap AIs default to
- multiple regressions in functionality (can’t delete a card easily, no 2FA check, a lot of purchases are missing because somehow they botched a migration?)
- they had a consistent design system for the FE before this, and now it looks like “Generic Gaming Website From 2025-2026. [Phrase 1. Phrase 2. Phrase 3]”
Even if they outsourced to devs that used AI, they are asking for a lawsuit to remove core security safeguards, while rolling old subscription plans onto the new plan in the same rewrite/migration.
Maybe it was different when you used it, but when I had Humble Choice years ago, I remember that it was very clearly communicated that the "pause" is for one month only. Maybe a dark pattern that you have to remember, but it was shown
To me it was always pretty clear. Though I have heard there was some probable weirdness in the functionality. Still bit questionable of grandfathered "classic" subscribers were forced to go through this to keep their status. But that is pretty normal, only have seen one site that allows coming back later to subs at old sub.
With them killing the "classic" it doesn't really matter anymore. Better just to cancel.
My order history is gone. The "new" look is bad, it seems another hallucinated neon brain fuck. Wasn't perfect before but why the need to change just for the sake of it? Why? What was improved? This is what pisses me off the most, using LLMs for doing things just for the sake of it
204 comments
[ 3.1 ms ] story [ 81.3 ms ] threadAI doesn't produce bad software, people do.
It's up to the people involved in the process, to gatekeep AI to not *ship* bad software.
You cannot just turn the saw on (if electric), point it at some wood and expect anything resembling furniture.
AI tools are complex systems that definitely can have agency, even though that is mostly human controlled. Bad controls = bad outcomes.
Yes, I totally agree that humans have the most control over them, but that doesn’t mean it’s completely off the hook.
If we really have to draw simplified real life comparisons, maybe a car or a gun would be more suitable “systems” that we can metaphorically compare AI to.
Does your IDE of choice produce bad software?
Know what changes? The amount of human effort needed to produce increasingly sophisticated software keeps falling. And know what doesn't change? People are still inputting requirements into the machine, testing what comes out, and iterating until it works as they'd like it to. That's the software production process in a nutshell. Doesn't matter if it takes a few thousand iterations over many months, or a single iteration that takes a few hours. It's still people who actually produce software, not the IDE nor the LLM (both of course generate code to help in the production process).
Before LLMs, only incredibly few programs were written by "inputting requirements into the machine", most code was about inputting instructions that fulfill requirements. Usually code is/was related to the requirements, but for anything non-trivial it was almost never a direct translation of the requirements.
So as far as I can see, you're trying to gaslight here.
Few programs were written in the past because the effort required was huge. Can you imagine how long it'd take to correctly code a simple calculator using only 0's and 1's? Or even hexadecimal? And then suddenly it took just a few lines of C. The transformation scale is similar now we can use natural language instead of having to learn a programming language.
> Few programs were written in the past because the effort required was huge.
What are you talking about? A lot of programs were written in the past.
The primary reason why programmers are/were needed is because humans and machines speak different languages natively, and someone/something needs to bridge the communication gap. That bridge is done via translating instructions/requirements into the many programming languages we've developed. Now we have LLMs to also do the translation.
> A lot of programs were written in the past
Yes, comprehensively, but also only "recently" (given any 2+ slices of advancement in tools used for software development). Somehow you missed the - I think - very clear "0's and 1's"/"hexadecimal" -> "C" and "programming language" -> "natural language" references.
I understand you perfectly well, I'm just calling out your misrepresentation of these words, because they are very much distinct in our industry, even today. Trying to say differently is simply gaslighting.
"Communication is commonly defined as the transmission of information. [...] A common way to classify communication is by whether information is exchanged between humans, members of other species, or non-living entities such as computers." - https://en.wikipedia.org/wiki/Communication
How did you find the page on communication, but not the one on programming?
What point are you trying to make?
AI makes producing bad software easier.
In particular they offer their owner a high maintenance decaying shelf life, or an explosion.
Of the 2,476 explosions to date, two were used specifically to kill people, some of the others did kill people, although such things were supposedly unintentional.
They have not afforded a better way to make canals, mine in a controlled manner, or fart a starship to space .. although people have desired such affordances.
Always review the code an LLM creates.
(When I told Claude that this is very bad, its reply was “Good catch!”)
It's always up to the skill of the human.
LLMs will be an interesting test of the current broken systems at most large companies. You can literally fire your entire engineering team and replace them with your "really smart" cousin and Claude. And that's going to be mighty tempting when the MBA mantra seems to be that cost reductions this quarter are infinitely more important than any potential costs in the future. But of course they're not.
But in this case, they seem to have skipped that as well.
If not, company is not viable and will crash badly soon. Market forces and all that.
No one at any company has ever understood every bit of the stack they ship; see all the bugs AI is finding in all the frameworks and libs we been shipping this whole time; oh yeah we sure knew what we were shipping.
git pull git@github.com:allofit/allthethings.git engineering patterns is what created this mess to begin with
I've seen people build great software but fail to evolve it.
All the lessons from agile, clean code, TDD and testable architectures, CI/CD, DevOps are really about building software that can evolve, yet most of those get ignored from time to time due to cost (frequently called time) pressures.
GenAI stresses this loop even more, and no human can effectively deal with the level of output you need to fit in to continue effectively shipping. This can only mean more bugs rather than fewer bugs.
Yes, in a well executed project, LLM agents can bring immediate value and find and fix bugs: because the system has the good feedback loop built in. As soon as you lose that (over shorter time with LLMs than with humans), it starts getting exponentially worse.
And this is all even before you stop and think if you are even building the right thing.
It’s exhausting, but so often, my mush-brain has to concede that it has a point.
On any given day, several hundred decisions in to what I’d imagined would be a sensitive but straightforward spec, I’m keenly tempted to just accept its default judgements…
By now, I’ve encountered the whole of all 3 official bodies’ standards in anger. For sure that’s not the same as real understanding or hard-earned experience. But I sure understand the formalities of the problem space a hell of a lot better than I did when I started this particular vibe coding adventure.
Contrast with my young friend whose approach was “make me a first person shooter, in Los Angeles, I want {some kind of weapon} and cars, make it super polished and fun”. And his follow-ups along the lines of “I don’t care, make it awesome, whatever’s most awesome.” I hate to say he’s holding it wrong (as he did get some sort of Unity/Godot thing out of it), but… does my neuroticism elevate my LLM-hectored coding out of “vibe” territory?
At the very least, if these are indeed different groups of morons, one group is shutting up whenever the other is spewing their flavor of bullshit.
Vibe coding is such an overused term now. Stats say what, 80% of programmers use AI to generate code? 90%? So everything is vibe coded. If you do well, no, it's still engineering. Someone makes a mistake? Cuts a corner somewhere? Vibe coded.
Given the sentiment that software is getting worse, buggier, and overall less stable, that isn't a comforting statement.
I am extremely skeptical that those numbers are even close to accurate.
Or more possibly the person incompetent one.
Because it allows you to produce so much code that it's not viable to carefully review each line/module, and get the benefits of the fast output.
Writing code yourself forces your brain to start creating pathways in your brain and you can verify the correctness of those pathways in your mind (e.g. am I actually doing this right, if I have seen such API would I be happy to use it,...)
Not saying we should write code manually, just showing the differences.
Programmers don't just write code. They also understand why they're writing the code and what it's supposed to do.
On the other hand, people incompetent enough to ship this couldn't have produced a working website if they had to handwrite the code, so in that sense LLMs are at fault.
Whether that says more about me or about LLMs These Days, I’m not sure…
I wondered why I did not get email code from them for login. Then it arrive one hour later... Seems like pretty big oversight...
I've been involved in many "rewrites". Almost exclusively, the result is worse for users at release.
Especially with "big bang releases". Where a team is working behind the scenes for months or longer, attempting to rewrite the old system "bug for bug". Releasing and then finding out that some users would "hold the space bar to make the CPU heat up, and configuring emacs to interpret that as 'control'" (XKCD 1172). First, users don't see any improvements for all the time that the new system was being built. Secondly, they get all the new stuff bundled. So while they may even get value from the rewrite, it is bundled with the stuff that makes it worse. Three new features, one new bug. And all everyone sees is that new bug.
But even rewrites where parts are swapped out (strangler vine pattern!) are worse for the users. They do get value fast. And in small increments. They can give feedback early. But the (much longer) period in which the replacements are implemented, is almost always one of confusion, complexity, and "having to learn new things every week". We all know the systems where "change email" is a link that leads to a form on some other site, in an old design rather than part of the "edit account" page. Because the "account service" has been extracted, but the "notification service" replacement on the backlog still.
"rewrites" are almost always for the benefit of the business, the technical team, or other internal reasons. If "benefit to the users" is a part of it, it's almost always secondary: so that we can build and release and scale new features and cool stuff for users faster than with the old legacy spaghetti debt.
> THEREFORE A COMPUTER MUST NEVER MAKE AN ARTISTIC DECISION
i thought i'd be hearing about downsides?.. elaborate plz
Is this actually illegal in a jurisdiction or against ToS of payment providers and they can get permanently banned from using card payments?
It was pretty frustrating not being able to download the things I'd paid for.
Without evidence, accusations like this are, ironically, just vibes-based.
I removed the automatic payment process from paypal so at least that's safe, I'd recommend others to do the same at this point.
Cannot wait for some insider to tell us the story of how this came to be
Maybe they should finally change name to Edgy Bundle.
Let me do 2fa when i have to pay not to just login to services i care very little.
And yes blizzard my wow account is not worth anything either.
Plenty of other players would strongly disagree. Blizzard pushes so strongly for 2FA because the support costs associated with people complaining about their WoW accounts getting compromised and drained are massive. If every player were to just shrug and start over from scratch, Blizzard wouldn't have bothered with it.
IIRC it was only via email, which is a PITA. I don't want to switch to my email every time I login.
Nope, it was also the regular time-based ones.
You’d be surprised; a friend of mine had a battle.net account under one of his emails, because apparently his email was used to make a WoW account in college and was banned. He had never played wow before, contacted support, and they didn’t say what happened but revoked the ban
I'm a subscriber to Humble Bundle Choice (their monthly offering), and it has been a pain to receive a code via email at every login.
I doubt you have to fight anything to get there.
Do we blame the bulldozer if it bulldozes the wrong thing? We're just not used to bulldozing yet.
The people who make the bulldozer don't market it as "intelligent."
And yes, I also blame the people getting on that bulldozer, bulldozing like there's no tomorrow. But mainly I blame the first two groups.
That metaphor has been stretched about as far as I can take it, but it's the tools and people getting access to tools they've not been trained on to do tasks they have no prior experience with. As someone who just laid laminate flooring myself for the first time, I'm keenly aware of both having the right tool for the job and the knowledge of how to use it and knowledge of how to do the task in general affects the end product. The mistakes I made are my own, not the cause of the incorrect tools I used at times.
Which is what your parent comment is doing. They didn’t say AI was to blame, they said this wouldn’t have happened without it. It’s likely your fault if you blow up your hands with a firecracker, but it’s also true that it wouldn’t have happened if firecrackers weren’t a thing.
> Do we blame the bulldozer if it bulldozes the wrong thing?
That’s such a tired analogy. By this point it’s honestly hard to believe it is made in good faith.
There isn’t a massive industry with tendrils in every facet of society trying to sell you on autonomous bulldozers that do the bulldozing for you without interaction, is there? Caterpillar is working on autonomous bulldozers, but even those aren’t targeted and advertised at every household as something you have to use.
AI has caused a generational set back on how stakeholders view the value of expertise knowledge, but the gun is still people, AI is just one of the bullets along with things like obsessive cost cutting.
My large orgs entire marketing strategy is now run by an MBA using Claude to generate 'idea and best practice docs'.
The process that lead us to not having a CMO or any other marketing leadership with domain knowledge? Consulting partnerships using MBAs.
The reason we're not changing any strategies even though we're burning money pper CPA? MBAs concerned seeking alignment and having the right narrative to change course.
Thr planned new course? More Claude.
The technology itself is not to blame, but maybe the marketing around it should be. AI coding tools are marketed as magic tools that can somehow spit out perfect solutions even if the input spec is garbage. Of course that is impossible, and no amount of technology can solve that problem unless every AI coding agent is equipped with guardrails that makes it stop if it gets the impression that its user is an idiot.
Thought experiment. A startup is at the end of their runway. Need to show income or go down. Businessmodel is to sell "one off" sales, and vast majority of their accounts buy once or twice (a year or ever).
What if they "accidentally" make it easier for hackers to take over those accounts. Accidentally, as part of a large overhaul or rewrite of their systems. Then, at least for a while, their accounts suddenly "buy" much more products per month or year.
Cynical, I know. But not unthinkable.
If they just focus on finding more fun bundles to sell and don't touch the site code again, ever, it will be fine. I have been happily buying things I usually do not need from them since Humble Indie Bundle 1.
People are rebuilding working things from scratch with AI just because they can.
AI eroded the greenfield advantage of rewriting a project from scratch because a rewrite generally meant that the code went through the fingers of the new developer. With AI that advantage is gone so you are better off incrementally improving the old code base to minimize overall maintenance costs and reduce the needed headcount as a result of this work.
It showed me that they just don't care anymore.
But it still baffles me as to how an LLM could make it this worse, when they already had decent security measures. No top-of-the-line model should've done this. I guess it can only be concluded that they didn't even care to spend some money on SOTA models, even DeepSeek 4.1 Flash, instead turning to the cheaper but also much less capable models to save a few pennies.
Even if they outsourced to devs that used AI, they are asking for a lawsuit to remove core security safeguards, while rolling old subscription plans onto the new plan in the same rewrite/migration.
With them killing the "classic" it doesn't really matter anymore. Better just to cancel.