Why focus on schemers in particular? (Sections 1.3-1.4 of "Scheming AIs")

This is sections 1.3-1.4 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 1.3 Why focus on schemers in particular?
• 1:14 1.3.1 The type of misalignment I’m most worried about
• 4:27 1.3.2 Contrast with reward-on-the-episode seekers
• 4:46 1.3.2.1 Responsiveness to honest tests
• 7:54 1.3.2.2 Temporal scope and general “ambition”
• 11:17 1.3.2.3 Sandbagging and “early undermining”
• 17:13 1.3.3 Contrast with models that aren’t playing the training game
• 23:13 1.3.4 Non-schemers with schemer-like traits
• 25:20 1.3.5 Mixed models
• 28:35 1.4 Are theoretical arguments about this topic even useful?