Cybernetica signs contract to develop cryptographically secure artificial intelligence with US Office of Naval Research

Pixellated image of a computer screen

“The project supports Cybernetica’s long-term goal to improve the security and privacy of data and AI systems for governments and enterprise alike.”

Dr Liina Kamm

Senior researcher, project lead for PAI-MACHINE

In June 2023, Cybernetica and the United States Office of Naval Research Global (ONR Global) signed a grant agreement for developing and speeding up cryptographically secure artificial intelligence (AI).

The PAI-MACHINE project will optimise algorithms for collaborative AI applications so that allies can share work together for the common good without having to share their confidential data in full. "We will build on Cybernetica’s experience in creating secure multi-party computation technologies, specifically the Sharemind MPC platform,” explained Dr Liina Kamm, project lead for PAI-MACHINE. "The project supports Cybernetica’s long-term goal to improve the security and privacy of data and AI systems for governments and enterprise alike.”

The project also strengthens international research collaboration between Estonia and the United States. "This project is the first ever project initiated and funded by ONR Global to an institution in Estonia,” explained Dr Mamadou H. Diallo, Science Director at ONR Global London. “Cybernetica’s research direction could bring a significant positive impact to building efficient secure systems for allied collaboration and we are looking forward to the results.”

The ONR-funded Synthesis of machine-optimised cryptographic protocols with applications in secure machine learning systems (PAI-MACHINE) research project will run from 2023 to 2026. The work will be conducted by Cybernetica’s Information Security Research Institute. The first experimental prototypes will be in the domain of healthcare, but future applications of the technology include speeding up implementations of novel cryptographic technologies, for example post-quantum cryptography.