Nothing, but nothing did before LLMs became a thing. The test lies in whether the submitter can explain the code they submitted, and more importantly, the why.