Opposing AI and data tools used in warfare and surveillance that treat people as mere variables. Technosolutionism:
The ASRG operates within a broader ecosystem of resistance to surveillance and AI extraction. It has been featured in prestigious European research projects like "Figure It Out," supported by the Creative Europe program, alongside partners such as Drugo More (Croatia), Labomedia (France), and the Unfinished Foundation (Malta). These collaborations help ground the group's radical tactics within the context of transdisciplinary workshops exploring art, science, and technology.
In recent years, the rapid advancement of Artificial Intelligence (AI) and Machine Learning (ML) has transformed numerous industries and revolutionized the way we live and work. However, as AI and ML become increasingly pervasive, concerns about their potential risks and vulnerabilities have grown. One organization at the forefront of researching these risks is the Algorithmic Sabotage Research Group (ASRG). In this article, we will explore the ASRG, its mission, and the critical work it is doing to identify and mitigate the hidden dangers of AI and ML.
: Developing methods for developers using static platforms (like Jekyll or Hugo hosted via GitHub Pages) to integrate scrambling scripts directly into their deployment pipelines. Contextualizing ASRG in the Wider Tech Resistance Landscape algorithmic sabotage research group asrg
The technical execution of algorithmic sabotage targets the weak points of machine learning architectures: , computational resources , and hosting infrastructure . Tactical Approach Target Objective Data Poisoning
: For high-stakes systems (e.g., algorithmic trading, drone swarms), the ASRG would research and publish “sabotage recipes” that any user could deploy. These would not be exploits requiring hacking skills, but physical interventions—a specific pattern of electromagnetic interference, a particular timing of button presses—that cause the system to shut down safely. The goal is democratic redundancy: ensuring that no algorithm becomes too powerful to be interrupted by the people it governs.
The Algorithmic Sabotage Research Group (ASRG) stands as a provocative and uncompromising response to the challenges of the digital age. As artificial intelligence systems grow more pervasive and their training data becomes increasingly valuable, the tactics of sabotage, data poisoning, and algorithmic disruption are likely to move from the fringes to the center of political and technological debate. Opposing AI and data tools used in warfare
The ASRG argues that this is a form of soft violence. The user is no longer a subject but an object to be sorted. The "black box" nature of these systems means that recourse is often impossible—one cannot appeal to a line of code. In this context, the ASRG identifies a vacuum of resistance. Where traditional activism might seek policy change, the scale and speed of algorithmic deployment often outpace legislation. The ASRG proposes a different approach: direct intervention at the code level.
The ASRG's research focuses on several key areas, including:
: Published openly across independent literary networks like Reincantamento Substack , this document presents sabotage as an act of solidarity that cuts through capitalist automaticity to build human social autonomy. These collaborations help ground the group's radical tactics
In the contemporary digital landscape, algorithms have ceased to be mere tools; they have become the architects of reality. They dictate what we see, what we buy, who we date, and whether we are deemed worthy of credit or employment. As these systems become increasingly opaque, powered by proprietary machine learning models and vast troves of personal data, the power dynamic has shifted decisively away from the individual. Enter the , a conceptual and practical collective dedicated to subverting, disrupting, and exposing the tyranny of automated decision-making. The ASRG posits that in an era of total surveillance and algorithmic governance, sabotage is not an act of destruction, but a necessary act of liberation.
Whether they are heroes, villains, or simply the first responders to a technological apocalypse depends entirely on which side of the latent space you stand.
When generative AI companies deploy web crawlers to scrape data without human consent, they burn significant infrastructure compute power. ASRG documents tactics to convert websites into algorithmic "tarpits".