PulseAugur
EN
LIVE 06:05:14

New AI technique processes encrypted data with improved performance

Researchers have developed a novel technique called Hash-based Homomorphic Artificial Intelligence (HbHAI) that allows AI algorithms to process data in its cryptographically secure form. This method utilizes key-dependent hash functions that preserve similarity properties, enabling existing AI algorithms to operate on encrypted data without modification. The approach offers significant performance improvements over traditional homomorphic encryption schemes and even processing on plaintext data. Key advancements include reducing the compression rate by up to a factor of 10, which decreases computation time and energy consumption, and the ability to arbitrarily reduce the final validation error of AI decision tests through the use of repetition error-correcting codes. AI

IMPACT This research could enable more secure processing of sensitive data by AI systems, potentially leading to wider adoption in fields requiring high levels of privacy and security.

RANK_REASON Academic paper detailing a new AI technique. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →

New AI technique processes encrypted data with improved performance

COVERAGE [1]

  1. arXiv cs.AI TIER_1 English(EN) · Eric Filiol, Jaagup Sepp ·

    Arbitrary Reduction of Validation Error for AI Decision Tests using Homomorphic AI and Repetition Codes

    arXiv:2606.28994v1 Announce Type: cross Abstract: This paper presents new results and breakthrough obtained with the HbHAI techniques (Hash-based Homomorphic Artificial Intelligence) proposed in \cite{filiol0,sepp}. HbHAI is based on a novel class of key-dependent hash functions …