The whole point of this is that people will generally put the least effort into work as they think they can get away with, and LLMs will accelerate that force. This is the future of how code will be "vetted".

It's not important whose responsbility led to mistakes, it's important to understand we're creating a responsbility gap.