aigosFHE

Lattice based encryption for vector databases

Align vector DB security posture with traditional encryption requirements for relational DB. Protect business sensitive and PII content.

  • Encrypt vector embeddings the same way traditional DBs are encrypted at rest

  • Full homomorphic lattice-based encryption are quantum safe

  • Computation over encrypted data allows for vector-based information retrieval in your AI/LLM pipeline

  • Wrapper over existing vector DB

  • Works with: Chroma, Weaviate, Milvus, Redis

Embedding vs Encryption

  • Avoid the common misconception that embeddings in vector DBs are secured

  • Full Homomorphic lattice-based encryption (FHE) keeps your data secured in a post-quantum era

  • FHE supports computation and information retrieval over encrypted vectors

Support for All Major Vector DBs

  • Continue to work on your existing system

  • AigosFHE adds a wrapper to your data ingestion and data retrieval pipeline

9 Key Elements for AI Audit and Why They Matter

AI audit refers to the evaluation of AI systems to ensure they work securely, without bias or discrimination, and are aligned with ethical and legal standards. While AI audit has existed for years, recent technological enhancements have triggered a new

Read More »

Securing Multimodal Language Models

We believe that the end-state for most AI systems will be multimodal. Security is key, especially in high exposure sectors like government and financial services. Securing a multimodal AI system is however challenging given input complexity, the breadth of potential

Read More »