Skip to main content

Think you can outsmart AI? Announcing ‘Behind The Mask’ – Our all-new cybercrime role-playing game | Play Now

Insecure Output Handling

« Back to Glossary Index

A risk facing AI systems when the output generated by AI models is mismanaged in a way that could lead to security issues, for example, exposing sensitive or proprietary information or causing unintended actions to occur

« Back to Glossary Index