# Manifestation

It is the process of artwork creation using the DNN of an AI Artist already created, trained and NFT minted. Let us take painting as an example for easy explanation. An AI Artist is a DNN consisting of multiple layers of neurons and interconnections connecting neurons in adjacent layers.

Each connection is associated with a weight parameter. Given a specific theme or caption, the neurons start processing in a layer-by-layer manner. Each neuron performs a summation over the element-wise product of input data and weight. The final layer will produce an output graphic result in the output layer, which is the AI Artwork created. Essentially, a DNN is a versatile mapping tool that can map any input to any output.

The training process tunes the weights to fit a desired mapping function to the target artistic style. In contrast, the creation process generates them according to the distribution reflected by the network topology and weight parameters.

Since DNNs have millions of weight parameters, they reflect a certain distribution and at the same time have a certain degree of randomness, which to some extent reflects the “free will” in the artistic creation. Figure 3 summarises the manifestation process.

![Figure 4: Manifestation process](/files/grGqwx0Ksi8emowcEvee)

After producing an artwork, it will be directly recorded on the Blockchain and stored in the distributed storage network, thus forming an AI Artwork digital asset.

![Figure 5: Summarised process of creating an AI Artwork](/files/U6inbHtfD5BllHEL2h5N)

When a MASTER or MAESTRO completes the creation of an artwork and uploads the artwork to the distributed storage on MATRIX, the mirror contract deployed on Ethereum will immediately execute the minting of the Artwork NFT on Ethereum.

The ownership information of this NFT will be determined by the corresponding DNN of AI Artist and those parties contributing the compute power and distributed storage in the artwork creation process.

![Figure 6: Minting of the Artwork NFT on Ethereum](/files/CiRN76fcHVsGEIR8niyw)


---

# 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://airtist.gitbook.io/product-docs/airtist-ecosystem/manifestation.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.