I've obsessed about language in software since I first read Domain-Driven Design.
Caring about the language in your system was never a luxury. It wasn't for "engineers with spare time". DDD covers a lot of ground, and it gets called over-engineered at times. Because of that, some folks felt skipping it was pragmatic and caring about it was theoretical.
Now everyone is discovering that vague language produces slop as their model reflects their systems vocabulary back at them. I'm seeing a lot of the same experiments, and people landing in the same place, but I'm not seeing a lot of people calling it DDD.
The investment was never about creating pretty architecture diagrams. It was about making intent explicit. Making the why of a process live in the language of the code, not in someone's head.
AI is just making the ROI more visible.
And, this keeps happening. Prototypes over specs. Test-driven development. Ubiquitous language. Feedback loops. Guardrails. Each time the industry rediscovers something it already knew, it arrives with the energy of something new.
A lot of AI advice is legit, but the chances are someone wrote a book about it twenty years ago.
That book is still worth reading.