Please comment if you experience this as well, but sometimes, AI just flat out does not work. Sometimes it’s brilliant, but there are times when even the most simple things absolutely don’t work. I hope someone who understands AI better than I can explain this to me, but what’s even stranger, it seems to happen across brands.
I was setting up a simple HTML page today. I wanted the logo aligned right and the button aligned left. Easy, right? I could have probably figured it out myself, but the whole point of this exercise is trying to give AI real world stuff to do so that I can have problems like this to better understand the platform. No matter what I did in Claude, it would not respond to my requests.
- I asked it several different ways.
- Gave it screen shots to demonstrate the problem.
- Loaded code where it was working correctly.
- Started a new chat with fresh code and instructions
No matter what, it gave me this:
or this:
I’ll link to my entire sessions in PDF below, but it wouldn’t work.
So… I jumped to ChatGPT
It got closer on the first try…
but it still took some responses to get it to work. I also don’t like their code implementation compared to Claude. It would just stop rendering code for some reason when it was in their split screen mode. I finally got it to work.
Something that someone that knew about web development would have solved pretty quickly.
An important lesson here is to have a “toolkit” not a tool. Don’t be afraid to switch if you are hitting a brick wall. People often ask me “what should I use” and unfortunately my response is generally, “What ever is working at this current time.”
If you know about these consumer facing overhyped, underperforming apps, your insight would be appreciated!
