SOURCE-LINKED INTELLIGENCE

ASCII Attack: Recontextualising Harmful Requests as Artistic Critique in Large Language Models

arXiv · AI, language, vision and robotics · article · Sep 2, 2026 · UTC

Safety alignment trains large language models to refuse harmful requests stated plainly, but that training is applied mostly to surface form. Requests that only recontextualise the same operational content, changing how the model reads it, are therefore only weakly covered. The ASCII Attack is one such recontextualisation. It is single-turn and black-box: one message, with no access to model internals. It embeds a fully legible harmful request in ASCIl-art characters, presents it as artwork, and asks for feedback. Unlike ArtPrompt, it hides nothing: the request stays readable. The reply is wri

Read original source ↗ Open in workspace

recordType

paper

region

Global