How will AI impact the future of CRUD apps?

In my experience, CRUD apps are complicated with business requirements and therefore bespoke business logic. AI will definitely assist with getting apps up and running but there will still need to be someone who takes the requirements, turns it into a functional product, updates it, maintains it and owns it.

I doubt it will be business/stakeholder people interacting with the AI. It could potentially be business analysts but I doubt they'll want to. It would be an addition to their current job.

That leaves the software engineers. Maybe a lot of software engineers will turn into _solution_ engineers or _product_ engineers. Their job will be to create the solution/product even if they're not writing code.

It won't.

What you've described is a bare bones CRUD app. The little I know about things like AI/LLM is you feed it text so it can learn. If the input is not good then the output is not going to be good either, doesn't matter what it does with the input.

We (as an industry) can't get feature requirements or business logic documented to be interpreted consistently by humans, who understand those problem domains in high fidelity, let alone some computers reading that text. If the translation of requirements to code isn't great, code to LLM to produce new code isn't going to cut it either.

Our industry jokes about all we do is CRUD apps but once an app is mature and beyond simple models, has integrations from a dozen APIs, has customers integrating via APIs, does reporting, needs to guarantee-ish transactions and most importantly is using derived sets of data for billing/invoicing it is much much more than a "CRUD app".

Lotus had their Approach database which was a wizard Q&A driven system for building basic db apps. For basic stuff it will be fine and will probably enhance the wizard driven stuff out there in basic tools like Excel and Access.

As the requirements become more complex, like having an email sent off after a Create or Update, some SQL stored procedure code needing to be triggered, at this stage because I havent seen enough of the dataset used for training to really know, but they could potentially know enough data to replace a large group of programmers.

What reduces my confidence in that thought, is things like Chap-GPT3 cant even get history right from a wiki page, and cant even code in a specific language.

Its generating pseudo code in some instances, so I think they have a way to go still.

Naah. People have been trying to disrupt CRUD for years already with all these no code tools but nothing comes close to building professional grade CRUD using real programming languages. For prototyping or very basic things we already have tools like bubble. We don't need AI

Every other day I get a new customer opening up with "Hey, I'd like you to build this super simple CRUD app for us, it's basically just a... ". And months later we're still publishing new code & updates because it's never "just this" and never "super simple". So I guess it's going to take a while until AI gets to this point.

Oh, and the other very popular request that we get is "Hey, we have an app built with and we need to re-build it with traditional ways because we can't get new features in it because the no-code platform doesn't support them..."

No impact. CRUD isn't that simple or it would have been automated out a long time ago, AI or no AI. Even low code or no code solutions have not been sufficient to disrupt this industry. There is often business requirements that must be modeled. Not to mention that not all aspects of these types of jobs are 100% focused on CRUD applications but sometimes small utility programs that move documents and data from one place to another. AI isn't at the point where they can learn to investigate how the business works, then ensure that the processes are unaffected as these processes are improved upon.

Similar to the way it currently works with AI generated video or even code. Yes you can create simple stuff just like on retool, but want it to look polished like or be extremely light weight and performant? Still going to have to go in and know enough to fine tune or create custom new components, backends, data models etc.

So basically a parking app that save you a spot, aka low hanging fruit now, sure the ai stuff will eat that market, but want something like Carrot Weather, ToDoist, etc you are going to need someone to build it.

Perhaps we'd be able to promptly enter data without predefined structure. AI will help suggest entry form structure progressively. Searching and reporting will be in natural language.

The source code output from ChatGPT looks right but that shouldn't be confused with is right.

Maybe the next generation of text AI models will do it :p

A team at work is maintaining a 30+ years “CRUD” application. The amount of complicated carefully maintained business logic the application contains is breathtaking. Yes it is a “CRUD” app but it would take many many years to recreate it. And look at SAP. It is also a “CRUD” application. But there is nothing simple about SAP.

In the short term, I don't think current-generation LLMs are particularly useful for CRUD apps.

I honestly have no idea what next-generation LLM performance will look like, but it could _possibly_ be used to generate entire simple CRUD apps.

In the long term this is a bit like asking how ICBMs will affect fortress wall construction.

CRUD only, wow, so easy!!! I'm pretty sure someone has seen the following joke:

The husband told by the wife: "Buy a watermelon, if you see eggs, buy a dozen" before going out. So he went home with a dozen watermelons.

Without the need for humans, it’ll just be “CUD”

It will be a commodity. Retool is already doing this anyways

I think AI will speed up my building of them.