SpaceComputer

Earlier this month, SpaceComputer advisor Dahlia Malkhi shared the most detailed breakdown of SpaceComputer yet, “Building Orbital Root of Trust with State-of-Art Technology.”  

The vision: sovereign computation in orbit that can survive even if Earth’s infrastructure fails, preserving trust-minimized systems for the future.

This thread distills her talk: from why space offers unmatched security, to the architecture and functionality, to where builders can get involved.

We build on a growing space ecosystem, where we can utilize global connectivity from Iridium’s Low Earth Orbit (LEO) satellite constellation, and an increasing number of providers enabling on-orbit compute devices (aka boxes or nodes).

Today’s reality: bandwidth is limited, latency is high. But within 10 years, these constraints will mostly disappear.

Web3 sales and marketing | Ideas are intangible without execution, so start now. berabaddies 🩷 | shefiorg S14 💫

There are several advantages to the security of space.
Once launched securely into orbit, a compute node becomes:

→ Tamper-proof: physically unreachable
→ Delete-proof: erased data is gone forever
→ Jamming-resistant: secure ground zones prevent interference
→ Geolocation-attested: predictable orbits provide Sybil resistance

Satellite-based software is not without vulnerabilities.

As of now, the root of trust in the supply chain is less mature, ground station communication, and upgrading software are attack surfaces.

Many of these are considered vulnerabilities since this is a nascent technology. With time, it will evolve.

We’re already exploring several use cases with partners and the community:

→ cTRNG (Cosmic True Random Number Generator)
→ Validator diversification
→ Proof-of-Location
→ Secure orbital co-processors
→ Immutable bulletin boards
→ Chain checkpointing to prevent long-range attacks

The most valuable use cases will be the ones we haven’t imagined yet.

As outlined in our Blue Paper, our architectural design goal is to minimize the trusted compute base (TCB) in orbit. Less software on the satellite itself means smaller attack surface, easier verification, and fewer upgrades. Workloads run in sandboxed environments, with scalability anchored to services on Earth.

Our software runs on a two-tier architecture model:

Celestial Layer 1: authoritative root of trust in orbit, optimized for limited and intermittent communication.

Uncelestial Layer 2: multiple high-throughput Layer 2 chains providing services on Earth, anchored to L1

For the Layer 2, we're introducing a third variation, a hybrid of commit chains and state channels that combine state-channel speed + rollup security.

The validators in the Layer 2s stake votes on finality, enabling near-instant confirmations on Earth while always deferring to Layer 1 for ultimate security.

We are not the first, nor will we be the last space company to turn what’s deemed impossible into an inevitable solution.

In 1958, the CORONA project solved the 'impossible' problem of returning satellite reconnaissance images to Earth before the invention of digital photography by dropping buckets of film from orbit to be caught midair by aircraft. It worked for over a decade.

In comparison, our challenges are easier, and our timeline to feasibility is far shorter.

We’re already collaborating with an open, evolving community. The Request for Proposal system is live on GitHub.

We’re building in the open. Join us, experiment with workloads in orbit, and help shape systems that will still be running long after Earth’s infrastructure has failed with 𝕤𝕡𝕒𝕔𝕖𝕔𝕠𝕞𝕡𝕦𝕥𝕖𝕣.

→ Watch the full video here: https://www.youtube.com/watch?v=x4EgWxrxF4o 
→ Read the full blog here: https://blog.spacecomputer.io/recap-building-the-orbital-root-of-trust-w-dahlia-malkhi/ 
→ Join the Community: https://t.me/SpaceComputerOfficial