Title

Algorithmic Sabotage Research Group (ASRG): Practical Framework for Detection, Mitigation, and Responsible Research

Related Work (brief survey)

  • Summarize relevant areas: adversarial ML, data poisoning literature, model robustness, anomaly detection, secure MLops, red-team/blue-team practices.
  • Identify gaps: operational detection in production, cross-system cascade effects, responsibility frameworks.

A registry of strategically offensive methodologies to destabilize AI-driven frameworks. ASRG Zine

  • Alert thresholds and prioritization heuristics (combine statistical significance with business-impact weighting).
  • Example detection rules (concise pseudocode snippets):

    Reclaiming Space: Moving away from "necropolitical" technologies that reinforce structural injustices.

    They threw a wooden shoe into the gears. The machine stopped. And no one got hurt.