[john_strinlai]

in a typical clean-room design, the person writing the new implementation is not supposed to have any knowledge of the original, they should only have knowledge of the specification.

if one person writes the spec from the implementation, and then also writes the new implementation, it is not clean-room design.

[post_below]

I believe the argument is that LLMs are stateless. So if the session writing the code isn't the same session that wrote the spec, it's effectively a clean room implementation.

There are other details of course (is the old code in the training data?) but I'm not trying to weigh in on the argument one way or the other.