Confidential computing since 20179 years of expertise

Confidential Computing

The PoCo Protocol lets applications be ordered, executed and paid for on data they never see, with hardware-enforced confidentiality and an onchain proof of every run.

Talk to us Start building

The problem

Running code on data has
always meant handing it over

Every cloud, every API and every model provider needs your data decrypted to be useful. For regulated workloads - patient records, trading positions, industrial secrets - that requirement has made most of them impossible to move. PoCo removes it: the code runs, the data stays encrypted, and the run leaves a proof.

Your data leaves your hands

Compute providers need your data in the clear. From that point on, whoever runs the machine can read it.

No proof of what ran

Off-chain compute returns an output and nothing else. No attestation of the code, the data, or the machine that executed it.

Chains keep no secrets

Everything a smart contract touches is public by design. Private logic and private state cannot live onchain.

The solution

PoCo removes the operator
from the trust equation

Proof of Contribution matches compute supply and demand onchain, runs the task inside a hardware-secured enclave, and validates the result against a consensus of independent workers. No data in the clear. No privileged operator. Full onchain verifiability.

DataProtector

Encrypt a dataset once and register it onchain. From then on you grant access app by app and user by user, price it if you want to, and revoke it at any time. The data itself is never exposed.

iApps in secure enclaves

Ship any code as an iExec app. It executes inside an Intel TDX enclave on an independent worker, on data it can compute with and never read. The worker cannot extract it either.

Proof of Contribution

Workers stake to contribute, results are confirmed by consensus, and payment settles in RLC. You get the output and the onchain record of how it was produced.

What you can build