Demonstrator Dataset: Archiving Hacktivism - Derivative Collections for High-Risk Publics

This demonstrator dataset presents a scalable, AI-assisted methodology for archiving politically sensitive, high-risk digital dissent. Preserving the traces of hacktivism—such as defacement screenshots, leaked manifestos, and protest media—poses significant legal, ethical, and infrastructural challenges. To address the binary "open vs. embargoed" access fallacy in traditional institutional repositories, this project introduces a reproducible workflow for generating safe "derivative collections." The dataset represents a processed subset of a time-bounded corpus focusing on three emblematic hacktivist campaigns: Anonymous’ Project Chanology (2008) WikiLeaks’ Cablegate reposts (2010–11) The Syrian Electronic Army’s defacements (2012) Methodological Innovation: The materials were processed using a custom multimodal prototype (developed via Google AI Studio) to bridge qualitative media studies with computational, AI-driven archival practices. The automated pipeline executes a rigorous triage process to: Identify Personally Identifiable Information (PII), copyright violations, and security risks. Generate precise redaction and watermarking parameters (surrogate creation). Automatically extract and structure metadata, mapping the unstructured subversive rhetoric to the Dublin Core schema and custom archival descriptors (e.g., Risk Category, Visibility Status). Dataset Contents: demonstrator_metadata.csv: The finalised tabular catalogue containing the enriched Dublin Core metadata and risk assessments for the pilot corpus. ai_outputs.json: The structured JSON arrays generated directly by the AI prototype detail the automated redaction plans and metadata extraction logic. pipeline_demo.mp4: A brief screen recording demonstrating the automated ingestion, analysis, and structuring of the hacktivist artefacts. This pilot dataset serves as a proof of concept, demonstrating that the preservation of volatile counterculture media is a collaborative, auditable, and scalable infrastructure problem—one solvable through careful technical design rather than blanket exemption from access.