Lattica, a Tel Aviv startup, has emerged from stealth mode with a $3.25 million pre-seed funding round to tackle AI's privacy challenge using Fully Homomorphic Encryption (FHE). This method allows data computation without decryption, thus safeguarding privacy. Their innovative approach, Homomorphic Encryption Abstraction Layer (HEAL), enhances FHE performance across various hardware. With rising concern for data security in AI and significant investor backing, Lattica is positioned to influence the landscape of privacy-focused AI solutions, particularly critical for compliance-heavy industries.
Artificial Intelligence has made significant advancements, but the challenge of processing sensitive data without exposing it persists, making privacy a crucial aspect of its development.
Lattica aims to address AI privacy challenges using Fully Homomorphic Encryption, which allows computations on encrypted data while maintaining its confidentiality, surpassing traditional methods.
FHE has been considered the 'holy grail' of cryptography, but Lattica's unique approach standardizes and accelerates FHE operations, enabling practical applications in AI computation.
The companyâs CEO, Dr. Rotem Tsabary, emphasizes that their solution is specifically designed for neural networks, combining hardware and software optimizations to enhance security in AI.
Collection
[
|
...
]