Go offline with the Player FM app!
The counting argument for scheming (Sections 4.1 and 4.2 of "Scheming AIs")
Manage episode 387983838 series 3402048
This is sections 4.1 and 4.2 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
1. The counting argument for scheming (Sections 4.1 and 4.2 of "Scheming AIs") (00:00:00)
2. 4. Arguments for/against scheming that focus on the final properties of the (00:00:32)
3. 4.1 Contributors to reward vs. extra criteria (00:01:08)
4. 4.2 The counting argument (00:03:42)
64 episodes
Manage episode 387983838 series 3402048
This is sections 4.1 and 4.2 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
1. The counting argument for scheming (Sections 4.1 and 4.2 of "Scheming AIs") (00:00:00)
2. 4. Arguments for/against scheming that focus on the final properties of the (00:00:32)
3. 4.1 Contributors to reward vs. extra criteria (00:01:08)
4. 4.2 The counting argument (00:03:42)
64 episodes
All episodes
×Welcome to Player FM!
Player FM is scanning the web for high-quality podcasts for you to enjoy right now. It's the best podcast app and works on Android, iPhone, and the web. Signup to sync subscriptions across devices.