Advancing Private AI Compute with secure, server-side memory
Introducing private, server-side memory to Private AI Compute for personal AI.
Get smart on it
Google DeepMind has developed a way to store AI assistant memory in encrypted cloud servers while keeping the encryption keys on a user's personal devices, combining cloud computing power with on-device privacy protections. Previously, cloud-based AI systems could not retain information between sessions, forcing a choice between privacy (limited to devices) or continuity (requiring cloud storage). This new approach uses hardware-protected "secure enclaves" in the cloud that temporarily decrypt data only when needed, then immediately re-encrypt it, allowing AI assistants to remember context across devices and time without exposing unencrypted information to anyone outside the user's devices. The announcement includes publication of technical documentation, a tamper-proof record of server software, and results from an independent security audit to allow external verification of the system's privacy protections.
Related news
A Coding Guide to TypeSafe AI Jev: Typed Decisions, Calibrated Confidence, and Speculative Fan-Out with a System One Model
In this tutorial, we work with Jev, TypeSafe AI’s first System One model, which does not generate text at all: we send it a piece of program state and a set of typed questions, and it returns choices, scores, and yes/no probabilities that our code can branch on directly. We install the official Python SDK, make a first call that uses all three question primitives at once, and look at how the shape of the state changes what the model can know. We then recompute the published confidence sta
NVIDIA Releases Nemotron 3 Diarization: A 100M-Parameter Open-Weight Model That Tracks 8 Speakers in Real Time
NVIDIA has released Nemotron 3 Diarization, an open-weight speaker diarization model on Hugging Face. It answers one question about any conversation: who spoke when. The 100M-parameter model tracks up to 8 speakers, including when voices overlap. One checkpoint handles both offline recordings and real-time streaming. Is it deployable? Yes. The weights are released under the OpenMDW License 1.1, which permits commercial use. It runs on Linux through NVIDIA NeMo, using Ampere, Ada Lovelace, Ho
