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

Joe Carlsmith Audio

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

บท

1. Varieties of fake alignment (Section 1.1 of "Scheming AIs") (00:00:00)

2. 1. Scheming and its significance (00:00:33)

3. 1.1 Varieties of fake alignment (00:01:29)

4. 1.1.1 Alignment fakers (00:02:30)

5. 1.1.2 Training-gamers (00:04:29)

6. 1.1.3 Power-motivated instrumental training-gamers, or “schemers” (00:10:14)

7. 1.1.4 Goal-guarding schemers (00:11:48)