# Data Utilization and AI Training

To prevent data users from leaking and disseminating valuable datasets, which would reduce their rarity and value, direct access to the data is restricted. Instead, the Alpha Chain introduces a trusted GPU machine supplier system.

**Trusted GPU Machine Suppliers**

Data users must submit their large models to machines operated by these trusted GPU suppliers. By utilizing the user's signature, these machines can access the fragmented data required for training through the Distributed Fragmented Multi-Party Computation (DFMPC) algorithm. This approach ensures that data remains secure and fragmented, preventing any single entity from reconstructing the entire dataset.

**Distributed Fragmented Multi-Party Computation (DFMPC)**

The DFMPC algorithm enables the secure combination of fragmented data pieces for AI model training. Each piece of data remains encrypted and fragmented, ensuring that even within the GPU supplier's machine, the data's privacy and security are maintained. The use of multiple parties in computation further enhances security by distributing the trust among several entities.

**Zero-Knowledge Proofs (ZK Proofs)**

The training process and the data used are verified through Zero-Knowledge Proofs (ZK Proofs). These proofs provide a mathematical guarantee that the training has been performed as claimed, without revealing the actual data or the model specifics. This method ensures that the training results are trustworthy and verifiable, maintaining the integrity of the data and the training process.

**Conclusion**

The Alpha Chain's approach to data utilization and AI training ensures that the integrity and security of datasets are preserved, preventing unauthorized dissemination and maintaining their value. By leveraging trusted GPU machine suppliers and advanced cryptographic techniques like DFMPC and ZK Proofs, the network provides a robust and secure environment for AI model training. This innovative framework supports the sustainable and secure use of high-quality datasets in the development of advanced AI models.


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.alphaos.net/whitepaper/alpha-chain/data-utilization-and-ai-training.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.