Safety, Ethics & Governance · intermediate · concept 171 of 176

⚖️Prompt Injection

An attack where instructions hidden inside content the model reads are followed as if the user had issued them. Direct injection comes from whoever is typing; indirect injection arrives through a fetched web page, document, or tool result, which makes it the central security problem for tool-using agents. No complete fix exists, so defences layer restricted permissions, separation of instructions from data, and human confirmation before consequential actions.