The entire discussion was centered around whether or not using AI to detect AI content would work, or if it would create false positives that harm human content creators.

It could work "well enough" for YouTube to consider it a success while still harming a fairly large number of content creators.