factual
Provable Guarantees Must Cover the Side Channel
We can already prove mathematical theorems that software behaves to specification—even for machine learning, showing performance within epsilon with high probability—and could prove a system always conforms to the uncertain-objective design no matter how good its learning components become; but a harder guarantee is needed that no circumstance lets the AI convince a human to overwrite its 'constitution,' analogous to security side channels like reading keystrokes via a laser on a window.
factualpending
Speaker
Stuart RussellEvidence Quote
“there is no circumstance under which the software would convince a human being to go in and overwrite the constitution, and that is much more difficult kind of guarantee to provide”
Source
94 | Stuart Russell on Making Artificial Intelligence Compatible with Humans— Sean Carroll's MindscapeCreated: 6/13/2026, 12:03:44 AM
Your notes
Loading notes...