A taxonomy of non-schemer models (Section 1.2 of "Scheming AIs")

This is section 1.2 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

Chapters

• 0:36 1.2 Other models training might produce
• 1:15 1.2.1 Terminal training-gamers (or, “reward-on-the-episode seekers”)
• 4:12 1.2.2 Models that aren’t playing the training game
• 4:50 1.2.2.1 Training saints
• 6:17 1.2.2.2 Misgeneralized non-training-gamers
• 9:22 1.2.3 Contra “internal” vs. “corrigible” alignment
• 10:15 1.2.4 The overall taxonomy