In hindsight, I think they were completely right and I feel kind of lucky that they drilled that in so much, because even into my mid 30's I don't have a ton of trouble or resistance to picking up new things. Sometimes I don't love the way new tech is going [1], but I still try my best to keep up with what's in demand in the industry (generally looking at job boards and looking at their keywords and making sure I have at least a cursory understanding of the stuff they're talking about). I will admit I don't completely love that AI is being used instead of junior engineers in some cases, largely because a lot of AI code is shit or flatout wrong in non-obvious ways, but I still have tried my best to utilize it and learn from it because it's clearly the way that things are going. [2]
I've been hired and lost/quit more desk jobs than anyone I know, and I attribute my ability to find work quickly to this characteristic.
[1] e.g. treating memory like it's infinite, disregarding CPU performance as a means of "getting more shit done", making configurations (arguably) needlessly complicated like Kubernetes, etc.
[2] For example, my latest project has been building an HLS and Icecast "infinite radio station" which picks a random song from my collection, feeds a prompt to OpenAI for DJ chatter in between songs,
As a higher income individual I conversely seem to have a lot of time to study and am not given a constant stream of work I must complete every moment. I also have the benefit of working from home and being able to spend a lot on training and upskilling.
When you are poorer you typically don't get this. The vast majority of your income is spent at the end of the week. Your job gives you zero time to explore and learn more. You likely commute and and may have a second job to make ends meet.
Just saying 'learn to code' here doesn't address the systematic issues.