The Algorithmic Sabotage Research Group (ASRG) studies how algorithms can be subverted, manipulated, or weaponized—intentionally or inadvertently—to cause harm to systems, users, and societies. ASRG’s work sits at the intersection of security, AI ethics, adversarial machine learning, and socio-technical policy. This post outlines ASRG’s core focus, research directions, real-world relevance, ethical considerations, and recommended actions for practitioners and policymakers.
The Algorithmic Sabotage Research Group (ASRG) is an anonymous, practice-led collective focused on "techno-disobedience" against the "algorithmic empire," defined by its 10-point manifesto. The group promotes "wildcat direct action" and "aesthetico-political" methods, including AI data poisoning and text-based traps to disrupt automated systems. Read the Manifesto on Algorithmic Sabotage at reincantamentox.substack.com . Drop #17. Manifesto On Algorithmic Sabotage
Moving beyond mere theory, the ASRG documents, curates, and develops actionable strategies to protect creators and digital infrastructure from predatory automation. A significant catalog of these tactics is compiled in their research index, . Sabotage Methodology Target Mechanism Operational Objective Data Poisoning Training Datasets algorithmic sabotage research group %28asrg%29
The Algorithmic Sabotage Research Group is part of a growing shift in tech critique that emphasizes over passivity . Instead of merely analyzing how technologies are harmful, ASRG invites practitioners to participate in the active, often subversive, disruption of those systems.
The central thesis of the ASRG is the reclamation of the term "sabotage." Historically associated with workers throwing wooden shoes (sabots) into machinery to stop production during labor strikes, the ASRG applies this logic to the digital sphere. The Algorithmic Sabotage Research Group (ASRG) studies how
Think of the 2010 Flash Crash, where a single sell order triggered algorithmic feedback loops that evaporated $1 trillion in 36 minutes. No code was "wrong." No hacker broke in. The system simply did what it was told, and what it was told was insane.
The technical output of the Algorithmic Sabotage Research Group centers on compiling and creating tools that allow individual web users, creators, and sysadmins to strike back against automated exploitation. Their documented tactics generally fall into three distinct pillars of active resistance: Tactic Category Core Objective Example Implementations The Algorithmic Sabotage Research Group (ASRG) is an
Perhaps the most biting critique is that ASRG members themselves possess the exact skills needed to commit algorithmic sabotage. A former member of the ASRG’s red team was banned in 2024 for selling a zero-day sabotage exploit on the dark web. The group acknowledges this risk and has since implemented psychological screening and blind-review protocols, but the shadow of the "reformed hacker" remains.