DESILO's 5th-Generation FHE Scheme 'GL' Recognized by International Academic Community -- Two Papers Simultaneously Accepted at IACR Crypto 2026

• Co-authored with FHE inventor Craig Gentry; simultaneous acceptance of both scheme and bootstrapping papers marks exceptionally rare academic feat
• Hundreds-fold speedup in matrix operations accelerates path to commercial Private AI

SEOUL, South Korea, May 14, 2026 /PRNewswire/ -- DESILO, a specialist in Fully Homomorphic Encryption (FHE) and Privacy-Enhancing Technology (PET), announced today that two papers on its 5th-generation homomorphic encryption scheme 'GL,' led by DESILO researchers, have been simultaneously accepted at Crypto 2026, the world's most prestigious cryptography conference.

Crypto is one of the three IACR flagship conferences (Crypto, Eurocrypt, Asiacrypt) and has been held since 1981. This year's edition will take place in Santa Barbara, USA, beginning August 17.

The two accepted papers are: the core scheme paper ("Fully Homomorphic Encryption for Matrix Arithmetic"), and a paper on 'GL Bootstrapping,' the procedure that enables unlimited computation in homomorphic encryption.

Having two papers from the same research line — the core scheme and its follow-up depth paper (bootstrapping) — accepted in the same Crypto proceedings is considered highly exceptional by the academic community. This signifies that the GL scheme has been recognized not merely as a theoretical proposal, but as a complete, independently validated 5th-generation FHE system.

The research was co-authored by DESILO researchers (Yongwoo Lee, Eunmo Yang, Youngmin Lee, and Hyojun Kim), along with Craig Gentry (currently Chief Scientist at Cornami), who first constructed a fully homomorphic encryption scheme in 2009 and subsequently received the Gödel Prize — the equivalent of the Nobel Prize in theoretical computer science. The bootstrapping paper also includes contributions from Cornami's Eric Crockett.

Homomorphic encryption is a next-generation technology that enables computation directly on encrypted data, maximizing security. As 'Private AI' — using AI safely without risk of exposing raw data — has emerged as a key industry theme, the importance of FHE has grown substantially. However, the slow processing speed due to intensive computation has remained a central challenge for commercialization.

The GL scheme offers a meaningful breakthrough to overcome this limitation. By designing a structure optimized for matrix operations — the backbone of AI/ML — the scheme achieves performance improvements of hundreds of times over 4th-generation technology in core computation areas. This is seen as a significant step toward practical commercialization of Private AI services that require large-scale data processing.

DESILO is also accelerating practical application work on the basis of these academic results. In April, the company integrated the GL scheme into its commercial FHE library DESILO FHE, leading the way in technical validation and productization for Private AI implementation.

Yongwoo Lee, Chief Scientist at DESILO, stated: "The simultaneous acceptance of both the core scheme and the bootstrapping papers means GL has been recognized on the world stage not as a single-paper idea, but as a complete 5th-generation FHE system. DESILO will not leave the GL scheme in the lab — we are steadily implementing it in our actual library and product stack, and we will lead the security paradigm of the coming Private AI era on the foundation of these academic achievements."

The DESILO FHE Library:
https://fhe.desilo.dev/latest/

Learn more about DESILO and Private AI:
https://desilo.ai

Media Contact
Howard Park
Co-founder / CSO
[email protected] 

About DESILO

DESILO is a deep-tech company advancing the future of Private AI through breakthroughs in cryptography. Specializing in Fully Homomorphic Encryption (FHE) and privacy-enhancing technologies (PET), DESILO develops infrastructure that enables AI systems to operate securely on encrypted data, allowing organizations to unlock the value of sensitive data without compromising privacy.

SOURCE DESILO