Ziyzhu (auto-discovered)

Personal AI should be built around the user's device, data, and model, rather than as a cloud service, to ensure privacy, freedom, and dependability. They envision an AI agent that acts as a local lay…

A new approach to agent safety, called deterministic agent safety, separates the LLM's reasoning from its actions by placing the model in a sandbox with no credentials or network access, and exposing …