News for hackers. Stuff that matters.

A comment on Peak LLM.

Re:Peak LLM by anyekwest on Sunday April 16, 2023 @03:27PM (#35593439)
What if we just train LLMs to remove prompt injections from inputs? I feel like this isn't an intractable problem.
[ Parent | on HN ]
Re:Peak LLM by te_chris on Sunday April 16, 2023 @03:39PM (#35593568)
The author addressed this: why would the model built on the hallucinating technique be able to police the main hallucinator
[ Parent | on HN ]
Re:Peak LLM by kolinko on Sunday April 16, 2023 @04:23PM (#35593930)
He didn't really.
[ Parent | on HN ]
Re:Peak LLM by TisButMe on Sunday April 16, 2023 @04:30PM (#35593990)
I now did in the parent comment :P
[ Parent | on HN ]
Re:Peak LLM by TisButMe on Sunday April 16, 2023 @04:29PM (#35593986)
(author here) How do you know what's a prompt injection vs actual content? If you train another LLM to tell you what's a prompt injection, how do you know it has 100% coverage of all possible injections? OpenAI has been battling people trying to bypass their prompt re-write filter, and as far as I can see, not really winning, just constantly adding stuff to their blocklist until the next thing gets discovered.
[ Parent | on HN ]