Prompt Injection

A technique — originally an LLM security concern — in which carefully crafted instructions embedded in a user prompt or referenced content override the model's intended behaviour, constraints, or safety rules. In accessibility research and practice, the term is increasingly used more broadly to describe user-side rephrasing tactics that push AI systems to provide information they would otherwise refuse, such as BLV creators rephrasing questions about a person's ethnicity or skin tone when the model declines to assume appearance details. Relevant to both the security posture of accessibility-facing AI tools and to the agency users exercise over AI-generated content.