aura@gts.foxsnuggl.es (@aura@gts.foxsnuggl.es)
aurelia but on gotosocial
ancient (32) sapphic transfem
non-idempotent
basic bitch, friend of corvids, fox
retired deadinsi.de admin and gdpr enthusiast
i make computers do my bidding for a living (who'd have thought)
if you're concerned about whether or not to follow, congratulations, your concern means you're fine
signal: https://signal.me/#eu/oKwV83-CvWp2mLgrrIIzeU-t9vP1NkHu67Ce106Qn6at7NOhY4tEJripnRTF8XuT
matrix: @aura:bundernet.org
Post
-
I wish that those surveys so often cited by InfoSec pundits that ask
Senza categoria@filippo the failure mode of LLM outputs is very different and harder to reason about, but ironically more often wrong in such an obvious way that it's easy to anthropomorphize the LLM and make false assumptions about future failure modes.
reviewing LLM output definitely requires a different type of vigilance.