Output-Invariant and Time-Based Testing – Practical Techniques for Black-Box Enumeration of LLMs

This site does not allow itself to be embedded. You can still read it on the original site — the toolbar below keeps your place in the directory.

In a recent pentest, I tested a system which automates and streamlines a time-consuming business process. The web app would process .docx business files by transforming unstructured data to structured JSON. To expedite the pentest (because time is precious), I asked the developer: "How does the backend parse the document? Is there any particular format, say, specific headings or column names?"…

Read on trebledj.me ↗