What is Fully Homomorphic Encryption? And why is it dubbed “the holy grail” of cryptography? Before we answer these questions, let’s dive into the definition of encryption itself.
Encryption is the process of turning plaintext data into ciphertext so that unauthorised parties have no access to its meaning. The parties with the encryption key can decrypt the data and read it in plaintext. There are several well-known encryption algorithms that have been approved for commercial use by the National Institute of Standards and Technology (NIST), for instance, RSA and AES.
While the encryption algorithms mentioned above are functional and secure for our computers’ current computational power, our ever-evolving world calls for enhanced security to keep up with the world’s most powerful computing innovation: quantum computers.
Quantum computers are powerful devices, allowing them to make numerous complex calculations in a very short time. With quantum computing on the rise, our security is at risk. Quantum computers possess the capability to compromise the confidentiality of widely used encryption algorithms. To put things into perspective, a quantum computer would enable a malicious third party to break a secure password in a matter of hours. With the current computational power, it would take millions of years to break a secure password! For this reason, a more powerful encryption scheme is needed to withstand this risk.
Homomorphic encryption (HE) is a groundbreaking innovation in the world of cybersecurity. HE enables the performance of calculations over encrypted data without the need for decrypting it first, ensuring that the data remains confidential through the stages of processing, transmission, and analysis.
The way is quite fascinating: plaintext is not directly processed by the encryption algorithm, but rather the calculations occur over the ciphertext. Since there is a mathematical relationship between the two, calculations over either the plaintext or the ciphertext will yield the same results.
The calculations being referred to in the context of HE are addition and multiplication. Encryption schemes that utilize only one of these operations are referred to as Partially Homomorphic Encryption (PHE) schemes, whereas ones that utilize both are Fully Homomorphic Encryption (FHE) schemes. Due to the underlying lattice-based cryptography, FHE schemes are quantum resistant, meaning that the tremendous power of a quantum computer cannot intercept data that has been secured using FHE.
FHE was first conceptualised in the late 70s by the inventors of the RSA algorithm, but due to the challenges posed by the complexity of this concept, the problem of constructing an FHE scheme remained unsolved. Thirty years later, specifically in 2009, Craig Gentry wrote his groundbreaking PhD thesis ‘Fully Homomorphic Encryption Using Ideal Lattices’ solidifying a construction for a FHE scheme, forever changing cryptography as we know it today. Due to its mathematical difficulty and immense power, FHE is hailed as the holy grail of cybersecurity by today’s cryptographers.
FHE is utilized in numerous applications where the confidentiality of data is crucial, such as protecting sensitive medical records, cloud computing, and banking, among others. As quantum computing advances and the demand for stronger security measures grows, FHE is becoming increasingly vital. While its widespread adoption is still developing, FHE holds the promise of revolutionizing how we handle sensitive data in the digital age.
Resources:
https://www.researchgate.net/publication/385876759_Encryption
https://www.cs.utexas.edu/~dwu4/papers/XRDSFHE.pdf
Homomorphic Encryption 101. Fully Homomorphic Encryption (FHE) has… | by Marc Joye | Zama | Medium