New Method Identifies Deepfakes by Facial Veins
amsterdam, vrijdag, 8 augustus 2025.
Researchers from the University of Amsterdam have developed an innovative technique to identify deepfakes by examining the unique vein patterns in the face. This method can help combat fake news and misinformation online, which is essential for media literacy and the reliability of information in our digital world.
How Does the New Technique Work?
The new method, developed by researchers at the University of Amsterdam, focuses on the unique patterns of veins in the face. These veins are as unique as fingerprints and can thus be used to distinguish real faces from fabricated deepfakes. By using advanced image analysis algorithms, these patterns can be accurately identified and checked for manipulation. This technique provides a robust way to detect deepfakes, even when they appear highly realistic at first glance [1].
Reliability and Applications
The reliability of this method has already been tested in various scenarios, including the verification of online videos and photos. The results show that the detection of deepfakes is significantly improved by analysing these vein patterns. This is crucial for combating fake news and misinformation, which are increasingly used to influence public opinion and sow social unrest. The technique can be applied in various sectors, such as media, politics, and law enforcement, to ensure the integrity of information [1][3].
International Collaboration and Innovation
In addition to the efforts of the University of Amsterdam, there are other international initiatives focused on detecting deepfakes. For example, researchers from UC Riverside and Google have developed a system called UNITE, which can detect deepfakes by analysing backgrounds, movements, and subtle signals, even when faces are not visible. These advanced methods underscore the importance of continuous innovation and collaboration to combat the growing threat of deepfakes [4].
Challenges in the ‘Arms Race’
While these new methods represent a significant step forward, the ‘arms race’ between AI creation and detection remains an ongoing challenge. As technologies for generating deepfakes become increasingly advanced, detection methods must constantly be adapted and improved. This requires continuous investment in research and development, both in academic and industrial settings. Additionally, there is an urgent need for education and media literacy to help individuals better distinguish what is real and what is not [1][3][4].
Media Literacy in Education
To address the challenges posed by deepfakes and misinformation, education and parents play a crucial role. Organisations such as KlasCement provide lesson materials and resources to raise students’ awareness of the functioning, applications, and risks of deepfakes. These materials form an important part of a broader lesson on media literacy, helping students think critically about the information they encounter online. The Media Literacy Network connects more than 1,000 Dutch organisations dedicated to creating a society where everyone is media literate [5].
Technological Protection
In addition to educational efforts, there are also technological solutions that safeguard digital identities. For example, Facephi Biometría, a Spanish technology company, has launched ‘Advanced Injection Defense’, an AI-driven add-on for their Behavioural Biometrics Solution. This technology detects and blocks deepfakes, device simulation, and manipulated recording channels with forensic-level analysis. It is a significant step in protecting digital identities and preventing fraud in sectors such as banking, healthcare, telecom, and the public sector [3].