What is Proof of eXecution?
Proof of eXecution (PoX) is a cryptographic, on-chain proof that your code ran correctly inside an attested Intel SGX enclave — verifiable by anyone, without trusting the operator.
Proof of eXecution (PoX) is a cryptographic verification mechanism that proves a computation ran correctly inside an attested hardware enclave. After your code executes on Ethernity Cloud, the network writes a signed proof to the blockchain — a permanent, public record that anyone can verify.
PoX answers the question: *"How do I know my code actually ran on real, secure hardware — and that the result wasn't tampered with?"*
In centralized cloud computing, the answer is: *"You trust the provider."* With Proof of eXecution, the answer is: *"You verify it on-chain."*
Why execution proofs matter
When you run code on someone else's hardware, you face a fundamental trust problem:
- Did the code run at all? — The operator could return a fabricated result without actually executing your code.
- Did the code run on secure hardware? — The operator could run your code on a standard (non-SGX) machine, exposing your data.
- Was the result tampered with? — The operator could modify the output before returning it.
- Can you prove any of this to a third party? — Even if you trust the operator, can you show an auditor, regulator, or counterparty that the computation was legitimate?
Traditional cloud providers solve problems 1–3 through legal agreements (SLAs) and reputation. But they don't solve problem 4 — there's no independent, verifiable proof.
PoX solves all four.
How PoX works on Ethernity Cloud
The Proof of eXecution mechanism is built into Ethernity Cloud's three-enclave architecture. Here's what happens when you submit a task:
Step 1: Task submission
You submit your code and encrypted data to the network. A smart contract matches your task to a suitable node based on resource requirements.
Step 2: Enclave execution
The node runs your task through three independent Intel SGX enclaves:
- TrustedZone validates your code and data checksums against owner signatures
- SecureLock executes your code inside an isolated enclave — even the node operator cannot read memory
- Validator independently validates the on-chain proof and unlocks payment
Step 3: Proof generation
After execution, TrustedZone generates a signed transaction containing:
- The result hash (cryptographic fingerprint of the output)
- A status code confirming successful execution
- Attestation data proving the enclaves ran on genuine Intel SGX hardware
Step 4: On-chain recording
This proof is written to the Polygon or Bloxberg blockchain via the _addResultToOrder smart contract method. Once on-chain, it's permanent, public, and independently verifiable by anyone.
Step 5: Verification
Anyone — you, an auditor, a regulator, a counterparty — can query the blockchain and verify:
- That your code ran on attested Intel SGX hardware
- That the result hash matches the output you received
- That the execution wasn't tampered with
- Exactly when the computation occurred
PoX vs other proof systems
| Proof system | What it proves | Used for |
|---|---|---|
| Proof of Work (PoW) | A miner spent computational effort solving a puzzle | Bitcoin, securing blockchains |
| Proof of Stake (PoS) | A validator has economic stake at risk | Ethereum, Polygon, securing blockchains |
| Zero-Knowledge Proofs (ZKP) | A statement is true without revealing the underlying data | Privacy coins, zk-rollups |
| Proof of Storage | Data is being stored correctly | Filecoin, Arweave |
| Proof of eXecution (PoX) | A computation ran correctly on attested hardware | Ethernity Cloud, verifiable compute |
PoX is complementary to these systems, not competitive. A blockchain uses PoS to secure its consensus. Ethernity Cloud uses PoX to verify that off-chain computations ran correctly and writes those proofs to a PoS-secured chain.
Use cases for verifiable execution
Regulatory compliance
Financial institutions need to demonstrate to regulators that specific computations — risk calculations, compliance checks, AML screening — ran exactly as described. An on-chain PoX provides immutable proof.
Research reproducibility
Academic researchers can prove that experimental computations ran on specific hardware with specific inputs, addressing the reproducibility crisis in science.
Multi-party computation
When two companies jointly analyze data, both parties need assurance that the computation ran correctly and neither party's data was exposed. PoX combined with SGX enclaves provides both.
AI auditing
As AI regulation increases, model operators may need to prove that specific inference or training runs executed on attested hardware without data exposure. PoX creates an audit trail.
Supply chain verification
Proving that data processing steps in a supply chain (quality checks, sensor readings, logistics calculations) ran correctly and weren't modified after the fact.
The technical guarantee
What makes PoX different from simply logging that a task completed is the hardware attestation chain:
- Intel's hardware generates attestation quotes that prove the CPU is genuine and the enclave code hasn't been modified
- The Validator enclave independently validates the proof and releases payment (a node cannot approve its own payment)
- The proof is signed cryptographically and written to a public blockchain
- The blockchain's consensus mechanism makes the proof immutable
This creates a chain of trust that goes from Intel's hardware, through independent attestation, to a public blockchain — without requiring trust in any single party.
To get started building applications with verifiable execution, see the developer documentation or explore the full three-enclave architecture.
