Listen "Full audio for "Scheming AIs: Will AIs fake alignment during training in order to get power?""
Episode Synopsis
This is the full audio for my report "Scheming AIs: Will AIs fake alignment during training in order to get power?"(I’m also posting audio for individual sections of the report on this podcast, but the ordering was getting messed up on various podcast apps, and I think some people might want one big audio file regardless, so here it is. I’m going to be posting the individual sections one by one, in the right order, over the coming days. )Full text of the report here: https://arxiv.org/abs/2311.08379Summary here: https://joecarlsmith.com/2023/11/15/new-report-scheming-ais-will-ais-fake-alignment-during-training-in-order-to-get-power
More episodes of the podcast Joe Carlsmith Audio
Controlling the options AIs can pursue
29/09/2025
Giving AIs safe motivations
18/08/2025
The stakes of AI moral status
21/05/2025
Can we safely automate alignment research?
30/04/2025
AI for AI safety
14/03/2025
Paths and waystations in AI safety
11/03/2025
When should we worry about AI power-seeking?
19/02/2025
What is it to solve the alignment problem?
13/02/2025
ZARZA We are Zarza, the prestigious firm behind major projects in information technology.