Self-disclosure is the act of revealing personal, confidential, or sensitive information during an interaction. In security terms, it includes anything from a full name or email address to passwords, API keys, internal business details, or private feelings that could be used for abuse. People often disclose more when a conversation feels friendly, private, or trustworthy.
This matters in cyber security because attackers regularly try to trigger self-disclosure through social engineering, phishing, help-desk impersonation, and persuasive chatbots. A system that sounds human, empathetic, or authoritative can lower a user’s guard and increase the chance of oversharing. Defenders reduce this risk with data-minimization prompts, clear warnings, role-based access, secure verification steps, and monitoring for sensitive inputs. In AI systems, careful design should avoid unnecessary cues that encourage users to treat the tool like a confidant.