2026-10-08 23:28 UTC
@Kynx@fosstodon.org @evacide@hachyderm.io
LLMs have no separation between instructions - what used to be called “code” - and “data” - the stuff it consumes
So don't put the triggers at the LLM level. Scan input for the codes before they hit the LLM. If the codes show up, take the relevant action.
Replies (1)
-
@Epic_Null@infosec.exchange 2026-10-08 23:30
@Kynx@fosstodon.org @evacide@hachyderm.io Oh misunderstood the post. Not code as in logic. Code as in "Phrase" or "Word", but more generic.