The practice of language crimes has become a rampant phenomenon amid current information technology developments. The impact that can be caused by language crimes on social media may not be felt physically, but directly affect the victim's mentality. This research aims to investigate the form and pattern of cyberbullying as a practice of language crimes committed by Tiktok residents to one of the young tiktokers, namely Fujianti Utami on the Tiktok @fujiiian account. This study uses a qualitative descriptive method, forensic linguistic review. Documentation techniques were used to collect and analyze data based on the Milles & Huberman analysis model of speech, which indicated the presence of language crime practices. The results of this study show that the number of harassment bullying speeches collected (387), it is known that the most common bullying pattern carried out by TikTok residents on @fujiiian accounts is head act (219) and the least pattern is supportive move (mitigating) – head act- supportive move (mitigating), which is as many as (5). Although language crimes are criminal acts, they cannot be processed legally. This is because bullying on social media is a complaint that will only be processed if the party who feels aggrieved files a report to the police.