> For the complete documentation index, see [llms.txt](https://docs.thetpa.ai/deai-tokenized-processing-architecture/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.thetpa.ai/deai-tokenized-processing-architecture/tpai-chain-full-architecture/trust-proof-of-compute-privacy-and-fake-tpas.md).

# Trust: proof of compute, privacy and fake TPAs

Users trust TPAI Chain because work is proven, data is protected and fake endpoints are removed by validators.

**Proof of compute, in layers.**

1. Every job produces a **signed usage receipt**.
2. Every job adds **hardware proof**: the GPU itself signs that the job ran untampered (NVIDIA confidential computing).

   What attestation proves: every GPU proves its hardware and approved serving software with NVIDIA attestation before it takes jobs, re-checked regularly. That attested software records fingerprints (hashes) of the exact model and adapter it ran, and of the request and the answer, in a receipt signed with the attested key. Confidential jobs also keep data hidden from the datacenter. External API jobs carry the owner's signed receipt, with no attestation.
3. **From launch, super nodes spot-check a random sample of jobs by re-running them; the sample is larger for new or low-reputation providers**. Every job records its model version, settings and random seed, so a spot check repeats it in deterministic mode and compares the result; where hardware causes tiny differences, the check compares the model's word-by-word probability scores against a set threshold.

   How cheating is caught and punished. After each batch of jobs is settled there is a challenge window in which any user or validator can challenge a job with evidence. A provider proven to have cheated (wrong model, fake receipt, inflated usage) loses part of its bond, refunds the affected users, and repeat offenders are removed from the pool. Whoever proves the cheating receives a share of the slashed bond. Penalties are set so that the expected cost of cheating is always higher than what cheating could earn. Exact sample sizes and windows are set after testnet. \[Preferred]
4. Every datacenter posts a **slashable bond**.

**Privacy, chosen per job.**

| Level        | GPUs                                      | Who can see the data                                                              |
| ------------ | ----------------------------------------- | --------------------------------------------------------------------------------- |
| Standard     | Any GPU                                   | Encrypted in transit and on disk; the datacenter could see it while the GPU works |
| Confidential | NVIDIA GPU in confidential computing mode | Nobody outside the protected GPU, not even the datacenter                         |
| External     | An outside API                            | The outside provider; never allowed for Confidential work                         |

Data never goes on chain: the chain stores only a receipt with a hash of the job. A TPA owner can set a minimum level, for example Confidential only for a bank.

**Fake TPAs are removed by validators.**

1. Any validator reports a TPA with evidence on chain.
2. Randomly chosen super nodes test it.
3. If confirmed, the TPA is paused at once; the owner's funds are untouched.
4. A 2/3 validator vote deactivates it and slashes the owner's bond.
5. The owner can appeal to the review council.
6. A correct report is rewarded from the slashed bond; a false one earns the reporter a license warning.

**The same process covers dApps.** Every dApp lives inside a TPA, so a harmful or fake dApp is reported, paused and removed the same way, and its TPA owner's bond is at stake.
