Part of Project LEO · Kadropic Labs
Orbital Distributed Verification
Satellite networks have operated on blind trust since the beginning of the space age.
No cryptographic verification. No mathematical guarantees. No way to know if a node in orbit is doing what it claims.
LEO ODV changes this. We bring the Zero-Knowledge Fragmentation layer of Project LEO to orbit —
delivering the first Byzantine-resilient, cryptographically verified intelligence layer for satellite constellations.
// The Problem
Every constellation in orbit today — navigation, communications, observation, defense — shares the same fundamental flaw. When a satellite computes a result and broadcasts it to the network, there is no cryptographic mechanism to verify the claim. The network trusts it. And that blind trust is a catastrophic attack surface.
A signal round-trip to geostationary orbit and back takes over 600 milliseconds. Real-time verification from the ground is physically impossible. Satellites have always operated without external oversight.
No existing satellite protocol provides cryptographic proof that a node actually performed what it claims. Compromised satellites can silently poison the network with false data — indefinitely.
Ground control stations, master clocks, central aggregators — every current architecture has chokepoints. One successful attack on a ground station can cascade across an entire constellation.
Next-generation satellites must process data in orbit. But updating AI algorithms mid-flight — with any guarantee of correctness — has never had a cryptographic solution. Until now.
// The Solution — ZKF
ZKF transforms classical zero-knowledge proofs from a single-prover bottleneck into a distributed attestation fabric. Each satellite proves only its local computation — not the global result — and Byzantine-resilient consensus reconstructs the global correctness guarantee across the entire constellation.
Each node's identity is bound cryptographically to its hardware. Replay attacks, node impersonation, and cross-round forgery are computationally infeasible. Verified via TEE attestation with three-tier fallback.
Full inner-product argument over Pedersen commitments proves correctness of local computation in O(log n) proof size. No trusted setup. No central verifier. Soundness rests only on the discrete logarithm hardness assumption.
Geometric median aggregation over verified fragments reconstructs the global result. The system tolerates up to one-third malicious or compromised nodes — with mathematical convergence guarantees for non-convex objectives.
// Applications
Each application below addresses a problem that has existed since the first satellite was placed in orbit. Each one has no cryptographic solution in any existing protocol.
Every satellite proves the correctness of its computations to its peers in real time — without a ground station. The network polices itself. Compromised nodes are detected and isolated automatically, with zero human intervention.
NAVIGATION · COMMS · EOEven if a hostile actor compromises one-third of a military satellite constellation — the network continues operating correctly. Not by redundancy. By mathematics. The Byzantine threshold is a hard guarantee, not a design goal.
DEFENSE · INTELLIGENCESatellites can update their own AI algorithms mid-flight, attaching a cryptographic proof that the update is valid and uncompromised. No need to transmit the full model. No need to trust the transmission channel.
EDGE AI · EARTH OBSERVATIONObservation satellites can share processed signals with the constellation without revealing raw imagery. Differential privacy at the node level makes statistical reconstruction of source data computationally impossible — even with thousands of intercepted transmissions.
INTELLIGENCE · SURVEILLANCENavigation constellations can prove the integrity of position and timing data to each other — without disclosing internal orbital parameters. A globally verified navigation network that trusts no single node and requires no external authority.
GPS · GNSS · TIMINGAGI systems will eventually operate at scales that make Earth-based centralization a bottleneck. Distributed orbital intelligence — self-verifying, self-evolving, cryptographically sound — is not a use case for Project LEO. It is the destination.
AGI · PROJECT LEO// Roadmap
Full Bulletproofs inner-product argument implemented. Byzantine-resilient ADMM convergence proved for non-convex objectives. Real Pedersen commitment prototype with zero simulation flags. Formal adversary model complete.
Adapting ZKF for orbital communication constraints. Defining satellite-specific fragment formats, quorum configurations, and degraded-mode protocols. Seeking Co-Founder with deep space systems background.
FPGA/ASIC co-design for sub-0.1ms proof generation. Full constellation simulation with injected Byzantine behavior. Partnership with satellite operators for ground-truth validation.
LEO ODV running on live satellite hardware. Verified distributed intelligence operating in orbit — the first step toward cryptographically sound AGI at planetary scale.
"The most important AI systems of the next century will not run on data centers. They will run on distributed orbital infrastructure — self-verifying, self-evolving, and impossible to shut down. LEO ODV is the first step toward that future."
— Bader Jamal Jabarin, Founder & CEO, Kadropic Labs
Module 9 — Project LEO Architecture
Project LEO is a complete decentralized AGI architecture — a cognitive mesh of autonomous nodes that learn, agree, and act collectively. ODV extends the ZKF security layer beyond Earth-bound deployments into the orbital domain, making space the next frontier of verifiable decentralized intelligence.