Undated
The DEFCON AI Village challenge McGregor describes was called the Generative Red Team Two, a change in form from the prior year.
“this was something called the Generative Red Team Two.”
Topic
DEFCON
5 facts from 1 episode, in date order. These are what speakers said on the shows, each linked to the moment it was said. They aren't independently verified.
Undated
The DEFCON challenge targeted a 7-billion-parameter open language model from the Allen Institute for AI, protected by a guard model placed in front of it.
“7 billion parameter large language model, along with its guard model, put in front of it”
Undated
McGregor says the DEFCON challenge required participants to show systematic weaknesses using statistics rather than single exploits.
“what we basically had to bring to the security world, which the security world does not like this, and I don't blame them for, is statistics.”
Undated
The most widely exploited weakness at the DEFCON challenge was the loosely configured handoff between the guard model and the underlying chat model.
“exploiting that handoff between the guard model and the underlying model was the kind of, you know, big vector that you could use that to print cash over the course of the competition”
Undated
McGregor calls reports of weaknesses found in testing flaw reports, because no one has been harmed, and wants flaw-reporting businesses like bug bounty systems for machine learning.
“which is flaw reports, because it's not an incident like no one's harmed.”
Episodes