I've seen a subtler variant of that. Since 2025 winter model improvements, low quality developers can now commit somehow decent stuff more regularly, but as before, they still don't own the solution space independently. LLM output is now larger than what they can envision and cracks will simply happen further away, only to be discovered by someone else later. It's strange everything changes so nothing changes phenomenon.
Most vibeslop these days is superficially good in that it has unit tests, compiles, passes the style guide, but it's bad on a macro level in that it makes the system more complicated, ignores the wider design of the app, implements what the prompter asked for and not what the app actually needed, etc.
I'm seeing a bad pattern of vibeslop being used to solve the immediate complaint of the user without stepping back and reconsidering things from a product perspective. Just add another conditional statement to make this very specific scenario the user complained about work the way they want.
