Niobium develops hardware accelerators designed to significantly boost the performance of Fully Homomorphic Encryption (FHE) software. This custom PCIe hardware supports all major FHE schemes, enabling organizations to process sensitive data while it remains encrypted. The solution makes Zero Trust Computing feasible by providing high-speed, low-power acceleration for encrypted computation.
Funding
$6.4M raised to dateRaised to date based on public sources. This may differ from the amount the company actually raised and is based only on what is publicly available on the internet.
Founders
Product
Problem
Processing sensitive data in the cloud or other shared environments exposes it to potential breaches and unauthorized access. Traditional encryption methods require decryption before processing, creating a window of vulnerability during computation. This limits the adoption of cloud computing and data sharing for applications requiring strict data privacy.
Solution
Niobium provides fully homomorphic encryption (FHE) hardware acceleration solutions that enable organizations to process encrypted data without ever decrypting it, ensuring mathematically guaranteed data privacy throughout the entire lifecycle. Their custom-built PCIe card accelerates the performance of FHE software by multiple orders of magnitude, making zero-trust computing feasible for a wide range of applications. The hardware accelerator supports all major FHE schemes and integrates with existing FHE libraries and software, offering flexibility and ease of deployment. This allows organizations to share and analyze sensitive data across organizational boundaries while maintaining its confidentiality and integrity.
Target Audience
Niobium's primary customers are organizations in sectors such as cloud computing, fraud detection, medical research, and AI that require mathematically guaranteed data privacy and zero-trust computing capabilities.
Features
- Custom-built PCIe card for FHE hardware acceleration
- Supports all major FHE schemes
- Integrates with existing FHE libraries and software
- Highly parallelized architecture for high-speed processing
- Low-power design for efficient operation
- Enables zero-trust computing for sensitive data applications