PROVENANCE FOR LICENSING OF ARTIST-SPECIFIC ASSETS USING NON-FUNGIBLE TOKENS (NFTs)
The technology disclosed relates to a provenance system for tracing licensing of targeted artificial intelligence (AI) systems. A training logic is configured to train an AI system on a training dataset that satisfies a target artist configuration by requiring that at least some training samples in the training dataset are sourced from a target artist, and to generate a trained version of the AI system. The target artist configuration characterizes a work of the target artist. The trained AI system is configured to construct an output that satisfies that target artist configuration. A deployment logic is configured to make available the trained AI system via a blockchain network. A provenance logic is configured to license the trained version of the artificial intelligence system to a plurality of licensees, and to validate that licensed copies of the trained version of the artificial intelligence system satisfy the target artist configuration.
1 . A provenance system for tracing licensing of targeted artificial intelligence systems, comprising:
memory storing a training dataset that satisfies a target artist configuration by requiring that at least some training samples in the training dataset are sourced from a target artist, wherein the target artist configuration characterizes a work of the target artist;
training logic, having access to the memory, and configured to train an artificial intelligence system on the training dataset, and to generate a trained version of the artificial intelligence system, wherein the trained version of the artificial intelligence system is configured to construct an output that satisfies that target artist configuration;
deployment logic configured to make available the trained version of the artificial intelligence system on a blockchain network; and
provenance logic configured to license the trained version of the artificial intelligence system to a plurality of licensees, and to validate that licensed copies of the trained version of the artificial intelligence system satisfy the target artist configuration.
2 . The provenance system of clause 1, wherein the provenance logic is further configured to assign a parent non-fungible token (NFT) to the trained version of the artificial intelligence system.
3 . The provenance system of clause 2, wherein the parent NFT is configured to be traced back from offspring NFTs that are assigned to the licensed copies of the trained version of the artificial intelligence system.
4 . The provenance system of clause 2, wherein the provenance logic is further configured to validate that downstream results of the licensed copies of the trained version of the artificial intelligence system satisfy the target artist configuration.
5 . The provenance system of clause 4, wherein the parent NFT is configured to be traced back from offspring NFTs that are assigned to the downstream results.
6 . The provenance system of clause 5, wherein the downstream results are digital assets generated using the licensed copies of the trained version of the artificial intelligence system.
7 . The provenance system of clause 6, wherein the downstream results are generated by processing new inputs through the licensed copies of the trained version of the artificial intelligence system, and generating new outputs that apply the target artist configuration to the new input.
8 . The provenance system of clause 7, wherein the new inputs are new images, and the new outputs are reconstructed versions of the new images overlaid with the target artist configuration.
9 . The provenance system of clause 5, wherein the downstream results are digital assets generated using further trained versions of the licensed copies of the trained version of the artificial intelligence system.
10 . The provenance system of clause 9, wherein the downstream results are generated by further training the licensed copies of the trained versions of the artificial intelligence system on a further training dataset that satisfies a further target configuration, and generating further trained versions of the artificial intelligence system, wherein the further trained versions of the artificial intelligence system are configured to construct outputs that satisfy a combination of the target artist configuration and the further target configuration.
11 . The provenance system of clause 10, wherein the downstream results are generated by new inputs through the further trained versions of the artificial intelligence system, and generating new outputs that apply the combination of the target artist configuration and the further target configuration to the new inputs.
12 . The provenance system of clause 11, wherein the further training dataset requires that at least some training samples in the further training dataset are sourced from a further target artist, wherein the further target configuration characterizes a work of the further target artist.
13 . The provenance system of clause 12, wherein the further target configuration conditions generation of the new outputs on one or more user-supplied parameters.
14 . The provenance system of clause 13, wherein the user-supplied parameters include a random seed.
15 . The provenance system of clause 14, wherein the random seed is based on a Mandelbrot set.
16 . The provenance system of clause 13, wherein the user-supplied parameters include a speech type.
17 . The provenance system of clause 13, wherein the user-supplied parameters include an art type.
18 . The provenance system of clause 13, wherein the user-supplied parameters include a current state of the blockchain network.
19 . A computer-implemented method of tracing licensing of targeted artificial intelligence systems, including:
storing a training dataset that satisfies a target artist configuration by requiring that at least some training samples in the training dataset are sourced from a target artist, wherein the target artist configuration characterizes a work of the target artist;
training an artificial intelligence system on the training dataset, and generating a trained version of the artificial intelligence system, wherein the trained version of the artificial intelligence system is configured to construct an output that satisfies that target artist configuration;
deploying the trained version of the artificial intelligence system via a blockchain network;
licensing the trained version of the artificial intelligence system to a plurality of licensees; and
validating that licensed copies of the trained version of the artificial intelligence system satisfy the target artist configuration.
20 . A non-transitory computer-readable storage medium impressed with computer program instructions to trace licensing of targeted artificial intelligence systems, the instructions, when executed on a processor, implement a method comprising:
storing a training dataset that satisfies a target artist configuration by requiring that at least some training samples in the training dataset are sourced from a target artist, wherein the target artist configuration characterizes a work of the target artist;
training an artificial intelligence system on the training dataset, and generating a trained version of the artificial intelligence system, wherein the trained version of the artificial intelligence system is configured to construct an output that satisfies that target artist configuration;
deploying the trained version of the artificial intelligence system via a blockchain network;
licensing the trained version of the artificial intelligence system to a plurality of licensees; and
validating that licensed copies of the trained version of the artificial intelligence system satisfy the target artist configuration.