The repository lists 24 stars, 2 forks, no open issues, and has no license listed. The README describes zkAI as a systematization of knowledge on trustworthy AI using zero-knowledge techniques, with a generated site covering verifiability and privacy across training and inference.
Collecting history — the radar snapshots this repo daily. The trend line appears after 3 days of data (1 so far).
What it is
zkAI: Verifiable and Private AI. A systematization of knowledge on making AI trustworthy, from zkSecurity. The "zk" is historical: zero-knowledge proofs are one technique here, not the scope. The full SoK is a generated site, one page per approach and one per paper, with every number rendered from the data and carrying its provenance.
How it works
Describes a 2×2 field framework mapping phases of model life (Inference, Training) against properties (Verifiability, Privacy). Provides links to:
- Proving inference: zkML, DeepProve, Jolt Atlas, zkPyTorch, zkGPT, zkLLM.
- Private inference: 2PC/MPC/FHE, Iron, BOLT, CipherGPT, Nimbus.
- Proving training: zkPoT, Kaizen, zkDL, Optimum Vicinity, ZKBoost.
- Private training: MPC/HE, PriFT, and verifiable-FL hybrids. Notes that the technique variety includes ZK proofs, MPC, FHE, TEEs, optimistic fraud proofs, or trace sampling.
Getting started
No explicit installation or usage commands are shown in the portion provided. The README references a generated site at a separate URL for the full content.
Recent releases
RELEASES (latest 0):
- none
Traction
Stars: 24 Forks: 2 Open issues: 0 Created: 2026-07-11 Last push: 2026-07-11
Behind the repo
No linked startup or company information is provided in the facts block.
Caveats
License: none listed Language: HTML License absence noted; no license information provided in the facts.