verifiable_encryption.md

Verifiable Encryption

^^^May 2025 overview talk – Slides here ^^^

A New Primitive Empowering Private Data Availability

“Don’t trust. Verify.”

This document introduces Verifiable Encryption (VE) and explores how it enables Private Blockspace - a transformative new primitive for accountable, offchain systems.

Verifiable Encryption

Traditional encryption ensures that ciphertext reveals nothing about the underlying plaintext. This guarantees privacy, but at a cost: you can’t verify anything about the data without decrypting it.

This creates a trust bottleneck - you must fully trust those who hold decryption keys to act honestly. But what if we could shift some of that trust to verification?

Verifiable Encryption (VE) makes this possible. By introducing constraints on:

• the plaintext,
• the encryption algorithm,
• and the keys used,

VE allows public verifiability of claims about the data without decrypting it. That means anyone can check that encrypted data meets specific criteria - without learning what that data is.

Data Availability

Many protocols rely on Data Availability (DA) for safety and liveness guarantees. In adversarial conditions - such as censorship or outages - it is critical that data be publicly available.

However, not all data should be exposed to the world.

Some datasets are too sensitive for full transparency. The challenge: How can we ensure critical data is available, yet only selectively disclosed under prearranged conditions?

The Power of VE

By combining Verifiable Encryption with Private Blockspace, we unlock a powerful new primitive: auditable yet private data.

With integration into existing or novel Key Management Systems (KMS), VE allows one to define:

• who can access decryption keys,
• under what conditions,
• and what can be verified without access.

This means:

• Anyone (users, smart contracts, offchain agents) can verify that encrypted data is available and satisfies certain properties.
• Only authorized parties can decrypt and access the sensitive contents.

Use Cases

We’ve outlined a few use cases below - but would love to hear your ideas too! 💡 Open an issue to share feature requests or novel applications of Private Blockspace.

Programmable Privacy for Web3 dApps

VE and Private Blockspace align closely with the principles of local-first access control, enabling secure collaboration across decentralized applications.

In a world where chain data is globally replicated and indexed, encryption at rest becomes essential for access control and selective disclosure.

Example Applications

• VE'd database for collaborative dApps with fine-grained access control.
• Private rollups with programmable cryptography, enabling obfuscated state.
• Private bridging and escrow sending verifiably correct but private messages around web2 and/or web3 apps.
• Drop-in support for existing DA users via a proxy service, simplifying migration to Private Blockspace.

Trustless Data Markets

With VE, Private Blockspace, and escrow contracts you can construct protocols to build trustless exchange of data access See the Stock0 media market hackathon project for some great inspiration!

Here is a diagram inspired by them of an example setup of inputs for a market:

flowchart LR
    Data["Data to be Sold"] --> zkVM_Algo["zkVM(transform media)"]
    zkVM_Algo -- "proven data transform w/ VE anchor" --> Contract["Marketplace on <dApp chain>"]
    Data -- "VE data" --> Celestia["Celestia"]
    Celestia -- "header" --> Blobstream["Blobstream on <dApp chain>"]
    Blobstream <-- "verify VE anchor and DA" --> Contract

Loading

Verifiable Private Backups

NOTE: Celestia does not guarantee that data will be available forever! See the docs on retrievability for the latest safe assumptions to use.

With Private Blockspace, sensitive data can be publicly published in encrypted form, with predefined methods for recovery - without revealing its contents.

This unlocks a new class of verifiable, resilient backups.

Example Applications

• Auditable storage: Confidential datasets can be verified to exist and be recoverable, while remaining hidden from the public.

• Disaster recovery: Critical encrypted data is guaranteed to be retrievable using known decryption methods, ensuring survivability without sacrificing privacy.

Example Architecture

The anchor acts as a bridge, connecting any protocol to a proof that some private data was made available.

Future Work and Research Directions

While VE for Private Blockspace is still evolving, the potential is enormous. Current implementations have limitations, but these are rapidly being addressed by:

• Enabling performance improvements, confidential compute, and scalable parallelization of Private Blockspace workflows.
• Hybrid systems combining:
  • Trusted Execution Environments (TEEs),
  • Multi-Party Computation (MPC),
  • and Zero-Knowledge Proofs (ZKPs),
• Account-centric key management systems empowering end-users to declare keys to use for VE by operators.

For more, see:

• (Historical) research document
• Account-centric model for Private Blockspace research discussion