Illustration | NEWS IMAGE
The language barrier within social media platforms is disappearing. Thanks to artificial intelligence (AI).
According to the information technology (IT) industry on the 13th, X·formerly Twitter introduced an automatic translation feature on the 7th. Using Grok, a generative AI from xAI, it automatically translates posts written in various languages, such as English, Korean, and Spanish, into the language of the user and displays them in the feed. Previously, users had to press a separate Translate button, but now it shows them as if originally written in their language.
YouTube, the largest video platform in the world, rolled out a multilingual auto-dubbing feature early on. With the full application of this feature last September, when a YouTube video plays, the audio in the video is automatically dubbed into the language of the viewer. This feature, which preserves the tone and emotion of the speakers, is based on Gemini, the generative AI from Google.
Beyond X and YouTube, many global platforms are introducing automatic translation. Reddit, the largest social community in the United States, has been expanding its AI-based automatic translation since 2024, and the short-form platform TikTok is also automatically translating speech in videos in real time through its auto-caption feature.
These changes result from the remarkable advances of large language models (LLMs). Unlike in the past, when translation functions were criticized as unnatural or error-prone, LLM-based translation understands complex context and nuance and is much more natural. This is why user reactions on X continue along the lines of “I did not realize the post was written in a foreign language.” Automatic dubbing on video platforms such as YouTube has to read nonverbal elements such as facial expressions and tone, and it became possible as LLMs were combined with related AI technologies such as speech recognition and synthesis.
Platforms are seen as seeking to activate communication among global users and, through borderless content consumption, to increase time spent and expand platform influence. In fact, when YouTube launched the auto-dubbing feature last year, it disclosed that more than 25% of average channel watch time occurred in foreign-language regions where the default language setting was not the original language.