'Artificial Fingerprints' Made from Salmon DNA... Eliminating Art Forgery Controversies

Research Team Led by KAIST Professor Donggi Yoon Develops Core Technology for Anti-Counterfeiting Platform

Domestic researchers have developed an anti-counterfeiting platform technology that makes physical replication impossible by simply brushing it onto high-value artworks using salmon DNA.

Schematic diagram illustrating a digital fingerprint formed using salmon DNA ink. Image source: KAIST

KAIST announced on the 23rd that Professor Dongki Yoon's research team in the Department of Chemistry developed a core security and authentication technology using random patterns generated during the self-assembly of soft materials.

With the recent development of the Internet of Things, various electronic devices and services are connected via the internet, enabling new functionalities while simultaneously advancing counterfeiting technologies that infringe on personal privacy, with frequent reports of related damages. Accordingly, there is a steady increase in demand for anti-counterfeiting technologies with stronger and higher security. Especially, as seen in the forgery scandal related to the portrait of a beauty by artist Cheon Kyungja, which took 30 years, additional burdens are placed on artists who are likely novices in the field of anti-replication. Instead of electronic methods to solve this, an optical method that is artist-friendly and forms a physical unclonable function (PUF) anti-counterfeiting platform technology that forms immediately when brushed on is needed.

Figure 1. Development of security technology using self-assembled structure maze formed on a substrate patterned with liquid crystal material. Photo by KAIST

The research team utilized the fact that random patterns spontaneously generated during the self-assembly process of two types of soft materials can serve as unclonable security features similar to human fingerprints. Even non-experts in security can implement anti-counterfeiting technology as if painting a picture. First, they used liquid crystal materials. When liquid crystal material is trapped inside a patterned substrate, spontaneous symmetry breaking of the structure occurs, forming a maze-like structure (Figure 1). Defining structures open to the right as 0 (blue) and those open to the left as 1 (red), these can be converted into digital codes (0 and 1) through machine learning-based object recognition, serving a fingerprint-like role. This is a revolutionary technology that does not require complex semiconductor patterns and can be observed with the resolution of a smartphone camera, making it accessible to non-experts. It has the unique feature of easily reconstructing information compared to existing semiconductor chip-based methods.

[Figure 1-1] A schematic animation illustrating the process of blue (0) and red (1) exploring the maze to form the PUF. Image source: KAIST

They also developed a technology using DNA extracted from salmon. When the extracted DNA is dissolved in water and brushed on, buckling instability occurs, forming random patterns similar to zebra stripes. These random patterns exhibit fingerprint features such as ridge endings and bifurcations, which can also be defined as 0 or 1 and digitized through machine learning. The research team applied widely used fingerprint recognition technology to these patterns, using them like artificial fingerprints. This method is easily producible with a brush and can incorporate various colors, making it usable as a new security ink.

The security technology developed by the research team uses only simple organic materials and a straightforward process, enabling low-cost and easy production of security codes. Moreover, it allows the creation of desired shapes and sizes according to the manufacturer's purpose, and even when produced by the same method, the randomly formed patterns are all different, enabling high security functionality and possessing immense market potential and possibilities.

Professor Dongki Yoon stated, “This research created patterns that play the role of human fingerprints, which even the manufacturer cannot replicate, by accepting the natural randomness occurring during self-assembly as it is,” and explained, “This idea can be the cornerstone of technology that applies the numerous random phenomena existing in nature to security systems.”

The research results were published in the international journal Advanced Materials on the issues dated May 5 and 6.

This content was produced with the assistance of AI translation services.