Weaker, more trusted AI systems can supervise and monitor more capable AI systems, creating a scalable approach to alignment where we leverage the capabilities of AI systems to assist in aligning more powerful AI systems.
normativepending
Speaker
Nick BostromEvidence Quote
“leverage the capabilities of these AI systems to assist in the alignment of AI systems... weaker AI systems that are more trusted supervising and monitoring more capable AI systems”
Created: 8/12/2026, 6:20:34 PM
Your notes
Loading notes...