South Korean police are developing a new tool to detect deepfake videos and images , which will be useful during criminal investigations and is expected to help tackle related scams that could target elections .
According to an announcement by the Korean National Police Agency (KNPA), the National Bureau of Investigation (NOI) will develop new software that will detect whether videos or files have been modified using deepfake techniques.
Unlike most existing tools that are trained on Western-based data, the model behind the new South Korean software was trained on 5.2 million pieces of data from 5,400 Koreans. It adopts “ the latest model artificial intelligence to respond to new types of video scams for which it was not pre-trained ,” the KNPA said.
See also: United Kingdom: Deepfakes will jeopardize the elections
The software will be able to detect whether AI technology and deepfake techniques have been used in a video in five to 10 minutes.
South Korean police said the tool has an 80% chance of detecting whether a video is authentic.
The deepfake detection tool will be directly integrated into investigative processes, however this data will be used to guide the investigation and not as direct evidence.
However, addressing the malicious use of Deepfake technology is essential. The ability to create convincing fake videos can lead to the widespread dissemination of information that is not true.
See also: Gold Pickaxe trojan steals your face for deepfakes scams
Additionally, Deepfake technology can be used to create images or videos that can cause serious damage to a person's reputation and image. This can have serious consequences, especially in cases where the content is used to blackmail or intimidate someone.
Fake videos can be used to distort reality, leading to serious political and social consequences, as people may be unable to distinguish reality from falsehood.
South Korea: A Deepfake Detection Tool Ahead of Elections
The parliamentary elections in South Korea are just around the corner, scheduled for April 10.
There have already been many attempts to spread disinformation and propaganda, with the help of artificial intelligence tools.
See also: Deepfake scam: How a fake conference call led to the theft of $25 million
🔒 Protect your privacy with Proton VPN
Swiss VPN from the creators of Proton Mail — strict no-logs policy, strong encryption, and built-in NetShield that blocks ads, trackers, & malware.
- ✔ No-logs, based in Switzerland (except 14-Eyes)
- ✔ NetShield: blocks ads, trackers & malicious domains
- ✔ Covers all devices — free version available
The link is an affiliate link — SecNews may receive a commission at no additional cost to you. It does not affect the independence of our article writing.
In February, South Korea’s National Election Commission (NEC) reported that it had detected 129 cases of AI-generated content (fake video and audio ) that violated a recently revised election law. The deepfake content was mainly spread on social media platforms and distorted speeches of some candidates to influence the public.
Source: www.infosecurity-magazine.com
