Collaborative generative artificial intelligence content identification and verification

A system for collaborative generative artificial intelligence (AI) content identification and verification, comprising:

• a computing system comprising at least a memory and a processor;
• a content verification subsystem comprising a first plurality of programming instructions that, when operating on the processor, cause the computing system to:
• receive input content from an external party;
• deconstruct the input content into a plurality of data segments using a fractal chunking process that enables scaled evaluation at multiple resolution levels, wherein the data segment can be analyzed at different sizes and resolutions for increasingly rigorous analysis ranging from coarse similarity to fine-grained analysis depending on use case importance and evidence requirements;
• for each data segment of the plurality of data segments:
• use a hashing algorithm to assign a hash value;
• add the hash value to a content group;
• compare each of the hash values of the content group or at least one of the group elements against a database of registered content elements or groups to identify a match or computed score, or meeting a specific rule-based criteria, wherein the registered content elements or groups comprise a plurality of hash values representing AI-generated content and human-created content or content of unknown origin;
• verify whether a plurality of hash values of the content elements or groups match any of the plurality of hash values representing AI-generated content or any of the plurality of hash values representing human-created content;
• when a match or a computed score or meeting a specific rule-based criteria is found with any of the plurality of hash values representing AI-generated content, identify the related plurality of hash values of the content group as AI-generated or label the content group as AI generated;
• and
• when a match or a computed score or meeting a specific rule-based criteria is found with any of the plurality of hash values representing human-created content, identify the plurality of hash values of the content group or graph-based entity or edge representations as not AI-generated;
• create vectorized representations of content or content relationship graphs or characteristic graphs from subgraphs relevant to each data segment or content piece, wherein the graphs support flexible node and edge attributes and vectors, and wherein the graphs support probabilistic weightings for the edges, and wherein the relationships include:
• sub-elements themselves, relationships between sub-elements, relationships of sub-elements to the content chunk as a whole, and relationships to other content including non-internet sources;
• perform vector comparisons between the subgraphs for multiple pieces of content using hierarchical recursive analysis at multiple scales;
• use the vector comparisons to further corroborate or dispel content similarity by considering:
• the input content itself;
• a distribution of the input content across multiple resolution scales enabling hierarchical recursive labelling and search;
• broader internet or other references to the content pieces;
• similar content based on the hashes or the graph relationships or other elements, including standalone content bits as well as similar content bits and identification of other entities using such similar bits of content;
• and
• temporal dynamics of the content pieces' utilization post creation, including provenance tracking of content propagation and usage patterns;
• and
• adjust the similarity score, AI or human generated content scores, or labels based on the vector comparisons and the considered factors.