South Korean lawmakers have taken a significant step in combating the spread of sexually explicit deepfake images and videos by passing a new bill. The legislation criminalizes the possession or viewing of such content, with penalties including prison terms and fines.
The move comes in response to a growing concern in South Korea over the proliferation of deepfake pornography in online platforms, particularly in Telegram group chats. These illicit deepfakes have sparked outrage and calls for stricter measures to address the issue.
Under the new law, individuals found purchasing, saving, or watching sexually explicit deepfake material could face up to three years in jail or be fined up to 30 million won ($22,600). Previously, the creation and distribution of such content were punishable by up to five years in prison or a fine of 50 million won ($37,900).
Once the bill receives approval from President Yoon Suk Yeol, it will come into effect, increasing the maximum sentence for deepfake-related crimes to seven years, regardless of intent.
South Korean authorities have been actively addressing deepfake sex crimes, with over 800 cases reported this year alone, a significant increase from the 156 cases recorded in 2021. Most victims and perpetrators of these crimes are reported to be teenagers.
Additionally, South Korean police have launched an investigation into Telegram to determine if the messaging app has played a role in facilitating the distribution of sexually explicit deepfake content.
Meanwhile, countries worldwide are grappling with the challenges posed by deepfake technology. In the US, Congress is considering legislation that would empower victims of nonconsensual sexual deepfakes to sue, as well as criminalize the publication of such content and mandate tech companies to remove it.
Earlier this year, a social media platform took action to block searches related to Taylor Swift after fake sexually explicit images of the singer surfaced on the platform, highlighting the urgent need for robust measures to combat deepfake content.