-->

You cannot sell AI written software

You may have seen it too. This trend of “I wrote some software to solve a<br>problem. I think it’s pretty great. Does anyone have any feedback?”. Maybe it’s<br>a budget app. Or some company management thingy, tracking sales. Or invoicing.

Maybe you take a look. It looks pretty slick. But then you get a feeling of<br>uncanny valley. It’s just not right. Maybe you can’t even put your<br>finger on it.

I’m not an accountant, so when I see some accounting software do something in a<br>different way, it’s interesting. Why is it that way? What can I learn from the<br>fact that a professional thinks it should be this way?

You already know what’s weird about it, if nothing else because of the title of<br>this post. The software works this way because the LLM wrote it that way.<br>There’s no reason. It’s not even wrong.

How do you give “feedback” on that? My feedback would be that you don’t<br>understand the problem you’re trying to solve, and have shown no sign you<br>intend to understand it, so how could you possibly think you can solve it?

You’re not asking for feedback. You’re asking for someone else to do<br>everything.

A house analogy

Analogies sometimes work, sometimes not. I’m not sure if this one will work.

You have a house with stairs that go nowhere, because “houses have stairs,<br>right?”. You didn’t put the stairs there, and you have no sense for why there<br>should be stairs in any particular place, so sure, why not there?

Contrast this with an actual architect. They know why they put the stairs<br>there. They know why they didn’t put the stairs on the other side of the room.<br>It’s not (just) that they have actual knowledge, it’s also that they actually<br>had to think about “how do you get to upstairs, and what’s the experience of<br>walking up those stairs?”. They had to make many small decisions, each of which<br>required thinking, and required that the choices add up to a coherent whole.

The AI put the stairs somewhere, and you said “sure, stairs”. They’re in the<br>kitchen, but since you only had to OK the stairs, not actively choose which<br>room they go in, it didn’t even occur to you that room selection was a thing.

You don’t need architect credentials to do this right. You could have selected<br>a room. It probably would have been a fine room. Maybe a professional could<br>give you feedback on that choice. But if you don’t make choices then you’re not<br>actually contributing to the thing you supposedly built.

You made no choices, so you don’t understand the problem you supposedly solved.

Jurassic park phrased it well

“If I may… Um, I’ll tell you the problem with the scientific power that<br>you’re using here, it didn’t require any discipline to attain it. You read<br>what others had done and you took the next step. You didn’t earn the knowledge<br>for yourselves, so you don’t take any responsibility for it. You stood on the<br>shoulders of geniuses to accomplish something as fast as you could, and before<br>you even knew what you had, you patented it, and packaged it, and slapped it<br>on a plastic lunchbox, and now you’re selling it”

Except you didn’t even “read what others had done”. Not even that bare minimum.

Artisanally crafted custom solutions

I use my own email client, written in artisanally hand crafted code,<br>like back in the stone age. It solves a problem for me. I have my own youtube<br>UI. And my own RSS reader. For a while I used my own SSH. I<br>know people who wrote their own shell, and that is their daily driver. I plan<br>to move this blog to my own webserver behind a custom SNI<br>router.

Those can get feedback. I’m not saying they’re great, but even if the feedback<br>is “why would you trust OpenSSL more than OpenSSH?”, it’s useful.

But honestly, if someone wants an email client kind of like mine, but<br>customized to their preferences, then they may be better off asking an LLM to<br>write their personal email client. That was not true 5 years ago. I don’t want<br>to use email threading. Maybe you do. It’d be annoying to support both.

AI can make useful custom stuff

I was having some glitches with my video conference, and wanted to<br>troubleshoot. I couldn’t find a tool that did what I wanted exactly.<br>gping came close, but not quite.

I wanted something that pings regularly and frequently, summarizing these many<br>pings about once a second, to illustrate subsecond network outages. And using<br>Linux ICMP sockets, so that it doesn’t require root.

I spent about as much time looking for the perfect tool as I then spent telling<br>an LLM to write it.

I had it write blipfinder, and it solved my problem. I could have<br>had this tool written for me before my meeting had even ended.

Two relevant points about this:

I am, actually, a network expert. I know right from wrong in this space.

I am not, however, vouching for this tool. It solved my problem, and it’s<br>amazing that custom throwaway tooling can be generated like this.

Just for fun, I had the LLM write it again. That version was<br>also acceptable. It made some different choices....