Arguably, it’s the LLMs that are doing the aping, and hence the ape-coding.

In that picture, aping is probably a step up from stochastic parroting.