normative
Check Reasoning Process Over Reinforcing Outputs
Because reinforcement has dangerous aspects—rewarding apparently-good behavior can train deception—it is more robust to verify the system's reasoning process and its understanding of ethics directly, grilling it to pull apart its understanding and having people continuously review its decisions and reasoning during deployment.
normativepending
Evidence Quote
“it's actually more robust to check the process of reasoning and check its understanding of ethics. To reassure ourselves that the system has a really good understanding of ethics, it should be grilled for some time”
Source
Shane Legg (DeepMind Founder) — 2028 AGI, superhuman alignment, new architectures— Dwarkesh Patel PodcastCreated: 6/13/2026, 3:31:07 AM
Your notes
Loading notes...