AI & ML Practical Magic

You can finally let an AI remember all your private files without the company that built it ever getting a peek at what it's searching.

April 6, 2026

Original Paper

Opal: Private Memory for Personal AI

Darya Kaviani, Alp Eren Ozdarendeli, Jinhao Zhu, Yu Ding, Raluca Ada Popa

arXiv · 2604.02522

The Takeaway

It solves the 'privacy vs. performance' tradeoff for personal assistants, making secure AI memory nearly 30 times faster than previous methods. This paves the way for truly private digital life-logs that are actually usable.

From the abstract

Personal AI systems increasingly retain long-term memory of user activity, including documents, emails, messages, meetings, and ambient recordings. Trusted hardware can keep this data private, but struggles to scale with a growing datastore. This pushes the data to external storage, which exposes retrieval access patterns that leak private information to the application provider. Oblivious RAM (ORAM) is a cryptographic primitive that can hide these patterns, but it requires a fixed access budget