cs.LGNov 10, 2026

Adversarial Risk in the Generative AI Era Necessitates Dropping the Small Epsilon Ball

Authors: Andrew C. Cullen, Neil G. Marchant, Paul Montague, Jiani Xie, Benjamin I. P. Rubinstein

Organizations: University of Melbourne · DST Group, Adelaide

Abstract

Adversarial Machine Learning research is currently centered on the search for the epsilon-ball: minimal perturbations designed to fool models while remaining imperceptible to humans. We argue that this fixation on ℓp\ell_p norms is a metrological trap that communicates a distorted perception of risk to defenders, one that is unaligned with real-world system vulnerabilities. This paradigm misleads defenders into optimizing for narrow, mathematical robustness at the expense of systemic security, while simultaneously incentivizing attackers to exploit non-overlapping alternative adversarial pathways. Ultimately, we argue that the dominance of the ℓp\ell_p ball does not just fail to reduce harm---it may actively facilitate it by communicating a false sense of security against the wrong threats.

Figures & tables

Appendix figures & tables5 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. Risk Under Pressure: Compute-Aware Evaluation of Adversarial Robustness in Language Models

    Jun 9, 2026Malikeh Ehghaghi, Boglárka Ecsedi, Marsha Chechik +1LLM AlignmentNeural Network Robustness

  2. Exposing Vulnerability Monoculture Through Systemic Adversarial Robustness

    Oct 19, 2026Leo Elmecker-Plakolm, Lukasz Szpruch, Avishek Joey Bose +1Adversarial RobustnessAdversarial Robustness of Language Models

  3. Robustness Cannot be Reduced to Regularization: Studying Adversarial Training Beyond the Linear Case

    Jun 19, 2026David A. R. Robin, Rafael Pinot, Yann ChevaleyreAdversarial TrainingTwo-Layer Neural Networks