The case for keeping AI reasoning monitorable
An essay coauthor challenges the idea that chain-of-thought reasoning will inevitably disappear, arguing that model designers can choose to preserve its monitorability.
TLDR
The ability to monitor chain of thought—the reasoning steps an AI model writes out—should be deliberately preserved, an essay coauthor argues. Sharing the essay as part of the DeepMind Institute launch, the coauthor says model designers have agency rather than having to accept that this form of reasoning will inevitably disappear.
Combined views
76K
10 Sources, first seen ago
