Loss of meaning in the translation of satirical content by TikTok’s AI-Generated Subtitles in @Dipalelu’s Account

Nabilah, Vania Shafa (2026) Loss of meaning in the translation of satirical content by TikTok’s AI-Generated Subtitles in @Dipalelu’s Account. Sarjana thesis, UIN Sunan Gunung Djati Bandung.

Preview	Text 1_cover.pdf Download (223kB) | Preview
Preview	Text 2_abstrak.pdf Download (107kB) | Preview
Preview	Text Surat Keterangan Lulus Plagiasi.pdf Download (301kB) | Preview
Preview	Text 4_daftarisi.pdf Download (213kB) | Preview
Preview	Text 5_bab1.pdf Download (196kB) | Preview
	Text 6_bab2.pdf Restricted to Registered users only Download (386kB) | Request a copy
	Text 7_bab3.pdf Restricted to Registered users only Download (317kB) | Request a copy
	Text 8_bab4.pdf Restricted to Registered users only Download (531kB) | Request a copy
	Text 9_bab5.pdf Restricted to Registered users only Download (177kB) | Request a copy
	Text 10_daftarpustaka.pdf Restricted to Registered users only Download (238kB) | Request a copy

Abstract

The development of artificial intelligence has changed how translation works, including on social media like TikTok. Many people now use AI-generated subtitles to reach international audiences. However, this technology still has problems when translating complex content such as satire. Satire uses irony, humor, and cultural references that are not easy to translate, especially by machines that translate word by word. This study tries to find out what strategies TikTok's AI uses to translate Indonesian satire into English and why meaning loss happens. This research used qualitative descriptive method with content analysis. The data were 51 subtitles from five satirical videos on TikTok account @dipalelu. These videos were chosen because each had more than one million likes and was uploaded between August 2025 and January 2026. The analysis used Gottlieb's (1992) theory about subtitling strategies and Newmark's (1988) theory about factors of meaning loss. The results show that TikTok's AI mostly uses transfer strategy. This means the AI translates directly without changing much. Transcription, decimation, and deletion strategy were also used by AI but not as much as transfer strategy. Strategies like paraphrase, expansion, and imitation did not appear at all. For meaning loss factors, translator incompetence is the biggest problem. This happens because AI cannot understand context, informal language, and satirical tone. Linguistic differences and cultural differences cause significant problems. This study concludes that TikTok's AI seems unable to translate satire content. The AI translates too literally and misses the hidden meaning, jokes, and cultural aspects. Content creators should check and fix AI subtitles before posting so that international viewers can understand the real message. Perkembangan kecerdasan buatan telah mengubah cara kerja penerjemahan, termasuk di media sosial seperti TikTok. Banyak platform kini menggunakan takarir berbasis AI untuk menjangkau penonton internasional. Namun, teknologi ini masih bermasalah saat menerjemahkan konten kompleks seperti satir. Satir menggunakan ironi, humor, dan referensi budaya yang tidak mudah diterjemahkan, apalagi oleh mesin yang menerjemahkan kata per kata. Penelitian ini berusaha mengetahui strategi apa yang digunakan AI TikTok untuk menerjemahkan satir berbahasa Indonesia ke bahasa Inggris dan mengapa terjadi kehilangan makna. Penelitian ini menggunakan metode deskriptif kualitatif dengan analisis isi. Data berupa 51 takarir dari lima video satir di akun TikTok @dipalelu. Video-video ini dipilih karena masing-masing memiliki lebih dari satu juta suka dan diunggah antara Agustus 2025 hingga Januari 2026. Analisis menggunakan teori Gottlieb tentang strategi subtitling dan teori Newmark tentang faktor kehilangan makna. Hasilnya menunjukkan bahwa AI TikTok kebanyakan menggunakan strategi transfer. Artinya, AI menerjemahkan secara langsung tanpa banyak mengubah. Strategi transcription, decimation, dan deletion digunakan oleh AI tapi tidak sebanyak strategi transfer. Strategi seperti paraphrase, expansion, dan imitation sama sekali tidak muncul. Untuk faktor kehilangan makna, ketidakmampuan penerjemah menjadi masalah terbesar. Hal ini terjadi karena AI tidak dapat memahami konteks, bahasa informal, dan nada satir. Perbedaan bahasa dan perbedaan budaya menimbulkan masalah yang signifikan. Penelitian ini menyimpulkan bahwa AI TikTok belum cukup baik untuk menerjemahkan satir. AI menerjemahkan terlalu literal dan melewatkan makna tersembunyi, lelucon, serta aspek budaya. Para pembuat konten sebaiknya memeriksa dan memperbaiki takarir AI sebelum mempublikasikan agar penonton internasional dapat memahami pesan yang sebenarnya.

Actions (login required)

View Item