Joe Carlsmith Audio
Audio versions of essays by Joe Carlsmith. Philosophy, futurism, and other topics. Text versions at joecarlsmith.com.
Joe Carlsmith Audio
Simplicity arguments for scheming (Section 4.3 of "Scheming AIs")
•
Joe Carlsmith
Use Left/Right to seek, Home/End to jump to start or end. Hold shift to jump forward or backward.
This is section 4.3 of my report “Scheming AIs: Will AIs fake alignment during training in order to get power?”
Text of the report here: https://arxiv.org/abs/2311.08379
Summary of the report here: https://joecarlsmith.com/2023/11/15/new-report-scheming-ais-will-ais-fake-alignment-during-training-in-order-to-get-power
Audio summary here: https://joecarlsmithaudio.buzzsprout.com/2034731/13969977-introduction-and-summary-of-scheming-ais-will-ais-fake-alignment-during-training-in-order-to-get-power
4.3 Simplicity arguments
4.3.1 What is “simplicity”?
4.3.2 Does SGD select for simplicity?
4.3.3 The simplicity advantages of schemer-like goals
4.3.4 How big are these simplicity advantages?
4.3.5 Does this sort of simplicity-focused argument make plausible predictions about the sort
4.3.6 Overall assessment of simplicity arguments