As of late 2026, the ASRG has reportedly turned its attention to large language models and generative AI. Their unpublished research (leaked via encrypted USB drives left in academic libraries) suggests that LLMs are peculiarly vulnerable to what they call —feeding an AI its own prior outputs in a closed loop until it produces nonsense or, more dangerously, produces perfectly persuasive lies.
: Promoting non-commercial, community-led IT infrastructures as alternatives to the "AI cloud". 📖 Recommended Resources
ASRG coordinates, documents, and analyzes strategies designed to undermine the reliable functionality of harmful AI frameworks. These tactics target different levels of the modern digital ecosystem: 1. Data Poisoning and Scrambling algorithmic sabotage research group %28asrg%29
: Evading corporate surveillance by feeding AI scrapers obfuscated or distorted content.
Injecting adversarial or corrupted inputs into a machine learning dataset. As of late 2026, the ASRG has reportedly
Asserting that the first step of technology is always political, specifically through radical feminist, anti-fascist, and decolonial lenses. 2. Strategic "Sabotage" Tactics
: A deliberate act of refusal or disruption against systems that prioritize profit and power over human dignity. Communal Constraint Injecting adversarial or corrupted inputs into a machine
Methods to corrupt data within AI workflows to undermine the reliability of the system.
: Developing a collective mentality to resist algorithmic violence and "fascist techno-solutionism." Related Entities (Potential Confusion)
It calls for dismantling "algorithmic domination" to create room for social autonomy and egalitarianism. Action-Oriented Solidarity: