Cloaked AI Demo: Part 1

This is a demo of Cloaked AI, the first and only encryption solution that gives you the power to protect sensitive data in your vector database of choice. In this demo, we’ll be walking through an example that uploads embeddings to Pinecone and modifying that example to encrypt the embeddings and queries before they go to Pinecone. This demo explains why this is important, how the encryption works, and how it impacts performance. The Jupyter notebook with encryption changes shown in the video can be found here: https://github.com/IronCoreLabs/ironc... 0:00 Introduction 0:10 Setting up Pinecone's semantic search example https://github.com/pinecone-io/exampl... 1:54 Vector database security risks and embedding inversion attacks 2:35 Intro to Cloaked AI and Alloy SDK https://github.com/IronCoreLabs/ironc... 3:20: How Cloaked AI encryption works 3:50 Update Jupyter notebook to encrypt embeddings and then query them in ‪@pinecone-io‬ using Cloaked AI 10:27 Accuracy and Performance https://ironcorelabs.com/docs/cloaked... 10:55 Cloaked AI License and Pricing https://ironcorelabs.com/products/clo...