Domains, auth, and databases are probably the stickiest software products of all time. I suspect it’s a combo of security concerns and the risk of eroding goodwill among developers (their most important customer base rn) if they move up the stack too quickly. Once you know that the goal is to become cloud vendors, train on/compete with their own customers, and own the entire software stack e2e it’s hard to really feel grateful that they’re dragging it out but planning on doing it anyway.
For those of us working in infra/SaaS outside these companies it’s pretty clear that the only viable path that doesn’t involve getting cannibalized is training your own agents/models. The new coding agent-infra-data business model is a path towards full commoditization and undifferentiated prompting in 3-4 vertically integrated walled gardens. If you ever start making real money on pure software infra they’ll just be able to eat you alive by implementing something similar, training on their own tools rather than yours, and integrating it into their stack (which all of your customers are on already).
Also, you should only ever use $20-200/mo subscriptions on work you want them to train against or you don’t mind automate yourself out of. Think a little bit about what you’re teaching them to do when you use their products, especially if they’re the primary interface you’re working in. People are going to start caring about this a lot when the AI companies feel safe enough to begin the “extinguish” phase of AI coding. Train your own models!
I do not see enough people discussing this and trying to position their businesses in a way to avoid destruction at the hands of the labs.
What can most even really do about it? Assuming they pull off what they're saying they will: it won't matter what you do, what vertical, what moat you think you have. It will be a concentration of capital and power we've never seen before, and frankly that's terrible for the world.
Currently what makes this too difficult for anybody but frontier labs is the lack of access to the full distribution of workloads/data they use to RL multiple separate envs/evals without regressing more than they advance in general capabilities. Because they are the primary buyers/builders of that stuff, they have no incentive or reason to allow anybody to replay or resample it but themselves. And it's all so very expensive to do in aggregate so there's not really demand for other product shapes except more openly available, collaborative/bundled RL that you could plug your specific workloads into (who the frontier labs would obviously not want to support with their business).
However, if someone were to build a collaborative rollout platform that you could use to train private workloads (ie create useful IP that doesn't just become profit for other labs/come from what they already can do), the upfront investment would be amortized over the very large number of potential buyers once it gets into the 5-8 figure range, who essentially have no other choice if they want to remain competitive in the technology industry.
Until openai/anthropic ipo the concentration of capital/spending and ndas/loss of employability is too concentrated for the best researchers to really do this without rocking the boat. And a lot of also-ran ai/saas have the same risk due to the lack of capitalized acquisition opportunities or AI vendors to partner with.
If the plan is ultimately to drive you out of business if you ever build anything profitable with their products, and monetize your knowledge without fairly rewarding or explaining their intent to do so, you might as well defect early and build what you inevitably would need anyway.
you could try to help build that!
There’s a good chance that the next step after open weight models is something that makes it easier to post-train existing models on new workloads without running an entire pipeline just for yourself. Somebody has to actually build it though.
On the other hand, death of a livelihood in a social structure that doesn’t care about you.
We’ve always had pump-and-dump trash. The terrible stuff loses and the good stuff wins. I don’t see why that would change.
The companies that ditch the management structures of yore and successfully build better software faster with the new tools will come out on top. I don’t think these companies will replace ALL saas subscriptions with in-house solutions, because one huge benefit of paying someone else to hold that bag is liability and regulatory overhead.
Vibe coding full circle
And it will be good enough for 90% of use cases.
I think it’ll just raise the automation bar that much higher. In other words, it will be good enough for 90% of yesterday’s use-cases. But tomorrow’s? I’m bearish.
I imagine the amount of onerous bureaucracy when interacting with any entity is going to explode if they expect you to do it in an automated fashion, for both private companies and the government. Especially as roles are cut because they can be “automated”. Except only 80% of the workload can be reliably automated, the other 20% requires a long tail of effort.
It is likely better in some ways and worse in others...and few of hte people involved will have the understanding to tell the difference.
- People in other industries actually have specialist knowledge (this is something IT people forget)
- AI is coming for those jobs too
- and so are lots of other people, not just displaced techies but from any number of other industries.
Thinking about it in advance should include training yourself up for other sectors.
If AI is successful, it will be a very different world of jobhunting. If it’s unsuccessful there will still be job losses for a while because of overspending, and because of the wider economic issues.
Software engineering on one end is going to be more like finance or medicine where the cost of poor performance is actually higher than the risk that they wont be worth what you pay them (ie there isn’t any amount of pay where it’s worth it to hire them). Then on the other it’ll be like knowing how to use PowerPoint. All the new tech categories have too many bottlenecks and challenges beyond just producing code to consolidate like that.
Are you going to trust a vendor who has been running for years with your business critical operations or brad_69's awesome saas skills and agents? Do you think these businesses are going to shift their whole product to be in Altman's walled garden when they don't need to?
When they have full DevOps, with multiple environments for testing, compliance with security best practices, can do database updates and data migrations, move the app smoothly between different platforms, streamline costs, and scale to at least hundreds of thousands (if not millions) of users without crashing and burning... then go ahead and worry. We aren't anywhere near that.
On the other hand, if you are purely a coder, and cannot run a project that does all of the above... maybe start your worrying earlier.
We aren't, but the AI Labs no doubt are. They just can't release it right now because their bread and butter is still developers burning tokens, so there is still a business problem to solve.
A team I was managing spent a whole year converting government forms into a web app. It took a while and there was a lot of discussion with the end users and a bunch of back and forth (keep in mind this is government software with a lot of onerous legal requirements). Then about 4 months ago, we just fed Codex the forms, along with the UI design and the legal requirements and it spat it out in 1 hour. Was it perfect? No. But our iterative loop became make the web app, show it to end users and then incorporate feedback and completely cut out the UX team, product etc. We were able to complete the conversion in 2 weeks with one guy (something like 150+ government forms/applications). For most simple to medium corporate development work it’s basically over.
I just don't see how we keep going like this without layoffs. The company cannot responsibly continue paying for this many engineers, when half, or less, could do the same job.
Like, why is it all still so buggy? Why does no website seem to consider 96 PPI displays in its design? Why don't we have ostensibly useful features like undo trees everywhere? Why is everything still a RAM hog? Why does the proportion of websites that pay no heed to accessibility seem to be going up rather than down?
I'm told we're living in the future, but it sure doesn't feel like it.
Like the old quip that a cup of coffee lets you do stupid things faster.
Flawless software isn't the goal, ever. Feature development is almost always the primary goal, and fixing bugs and performance issues is only done if the bug is bad enough.
Lack of information. When you are on the building side you become blind to its faults. Consumers almost never tell you what isn't good about it.
> why is it all still so buggy?
Also lack of information. I haven't seen any bugs in years in the things I've worked on. LLMs have closed every last obvious testing gap, squashing any outstanding bugs there. That's not to say it is all bug-free now, but what bugs users may be experiencing aren't being reported.
AI is good at building software, but it is still not good at consuming software like actual consumers do.
I have been the lone developer on a project for quite some time and was already feeling that way. Yet, recently found out that a second is being added to the team and a third is to come. I guess I'll enjoy what is effectively a paid vacation before it all comes crumbling down.
Yesterday my dot made a web app to view 3D microscopy files
https://mucus-atlas-viewer.pazimzadeh.chatgpt.site/?sample=k...
I didn't even know about the sites feature
On top of it we added bacteria (green) and bacteria sized beads (red). On the bottom are colon lining cells (blue).
The inner mucus layer of the colon was generally thought to be impenetrable to bacteria, serving as a way to protect you from the trillions of bacteria in the gut.
It turns out that actually, it’s really easy for bacteria to swim through mucus by using flagella. We found some bacterial mutants that swim through mucus (the green ones). The red beads serve as a control marking the top of the mucus since they can’t get through the mucus.
Find this and other images in my paper: https://pmc.ncbi.nlm.nih.gov/articles/PMC11784811/
The real question is then - if they can get through the mucus at any time, why don’t they? And there are various theories about this - which I’m happy to elaborate on this you want. The basic idea is there is a risk whenever you cross the mucus (you get closer to the immune system) and mucus actually feeds and tames bacteria over time by providing them with various sugars and things like that. But if they get desperate/starve, then they will cross.
Mucus is still a mystery - there are substructures within mucus, take a look at this image for example:
https://mucus-atlas-viewer.pazimzadeh.chatgpt.site/?sample=f...
And here is a good paper too from my collaborator (who actually took these images) who discovered a sub-type of mucus and mucus-producing cells:
Once the app gets a bit more complex, it won't be that much different to any more traditionally vibe coded application (code, git/pr, test, deploy, ops). It only removes the barrier to get started, but doesn't solve any of the real problems. Problems like AI slop, code rot, bad architecture, security issues, data loss/breach because of bad code, and so on.
https://www.stripeeconomics.com/p/the-saaspocalypse-was-more...
* Some people do think this (e.g., https://sockpuppet.org/blog/2026/09/25/what-even-is-an-os-no...), but I remain somewhat skeptical that this'll be true for the most economically significant applications. This is not because I'm an AI capabilities skeptic (I'm not, I'm more of a doomer), but because even in a world where AI is designing and implementing all the software, there are probably still significant benefits to standardization. I see it as sort of analogous to compilers; we might get a world where compilers are both used and maintained almost exclusively by LLMs, but we probably won't get one where compilers stop being used or maintained at all because LLMs write everything in assembly.
Also I dont get the inhouse argument, you dont really save anything. So instead of paying X to SaaS, you now propose to pay X to anthropic instead AND maintain said SaaS... what? how is that better, that just shifted the cost to another company and you got the burden of maintenance and responsibility.