Joe Carlsmith Audio

Varieties of fake alignment (Section 1.1 of "Scheming AIs")

November 16, 2023 Joe Carlsmith
Varieties of fake alignment (Section 1.1 of "Scheming AIs")
Joe Carlsmith Audio
More Info
Joe Carlsmith Audio
Varieties of fake alignment (Section 1.1 of "Scheming AIs")
Nov 16, 2023
Joe Carlsmith

This is section 1.1 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

Show Notes Chapter Markers

This is section 1.1 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

1. Scheming and its significance
1.1 Varieties of fake alignment
1.1.1 Alignment fakers
1.1.2 Training-gamers
1.1.3 Power-motivated instrumental training-gamers, or “schemers”
1.1.4 Goal-guarding schemers