instruction hierarchy
1 statements across 1 episodes · 1 bullish · 0 bearish · 1 people on the record · first statement Aug 15, 2025 by Greg Brockman · across every show →
Everything said about instruction hierarchy, oldest first
Aug 15, 2025 positive
Brockman: Instruction hierarchy orders trust by system, developer, and user
“With instruction hierarchy, you sort of indicate that, hey, there's this message is from the system. This message is from the developer. This message is from the user and that they should be trusted in that order. And so that way the model can know something t…”