Under the law, AI-generated content must be labeled as such, and platforms may take action against content that carries no such label. In one case, a user's self-created content was mistakenly identified by the platform as undeclared AI-generated material and was consequently hidden, while the user's account was muted. After appeals proved unsuccessful, the user brought a lawsuit in court. The platform declined to explain its identification mechanism, citing technical secrets and algorithmic logic. The court found that the platform had failed to provide reasonable evidence or explanation for its determination, that the handling in question lacked a factual basis, and that it constituted a breach of contract. The court ruled that the platform must restore the hidden content and delete the records of the violation. This case makes clear that when platforms use algorithms to identify AI-generated content and take action on that basis, they must give users an appropriate explanation of the grounds for their determination and cannot refuse to explain solely by invoking trade secrets. It also recommends that platforms establish a tiered handling mechanism, apply the principle that penalties should fit the offense, and improve their content review and wrongful-determination relief mechanisms.