Sumerians

@Lagrange Official DeepProve has been establishing new rules for AI to be 'trustworthy', and the new direction revealed on July 31 is even more interesting, specifically addressing why AI thinks the way it does through reasoning proof, transforming the AI's black box thinking into traceable logical credentials.
Currently, using AI can be quite confusing: one knows it has provided a result but cannot fathom the underlying logic. For example, if a medical AI says 'this is a benign tumor', why does it make that judgment? What is the basis for the drone in a defense system locking onto a target? Previously, it was impossible to delve deeply into these questions.
But DeepProve's reasoning proof can achieve: using encryption technology to break down the AI's decision logic into 'verifiable evidence' without exposing the model's internal parameters or user privacy data.