Podcast Episodes
Back to Search“Accountability Sinks” by Martin Sustrik
Back in the 1990s, ground squirrels were briefly fashionable pets, but their popularity came to an abrupt end after an incident at Schiphol Airport …
1 year, 5 months ago
“Training AGI in Secret would be Unsafe and Unethical” by Daniel Kokotajlo
Subtitle: Bad for loss of control risks, bad for concentration of power risks
I’ve had this sitting in my drafts for the last year. I wish I’d been …
1 year, 5 months ago
“Why Should I Assume CCP AGI is Worse Than USG AGI?” by Tomás B.
Though, given my doomerism, I think the natsec framing of the AGI race is likely wrongheaded, let me accept the Dario/Leopold/Altman frame that AGI …
1 year, 5 months ago
“Surprising LLM reasoning failures make me think we still need qualitative breakthroughs for AGI” by Kaj_Sotala
Introduction
Writing this post puts me in a weird epistemic position. I simultaneously believe that:The reasoning failures that I'll discuss are st…
1 year, 5 months ago
“Frontier AI Models Still Fail at Basic Physical Tasks: A Manufacturing Case Study” by Adam Karvonen
Dario Amodei, CEO of Anthropic, recently worried about a world where only 30% of jobs become automated, leading to class tensions between the automa…
1 year, 5 months ago
“Negative Results for SAEs On Downstream Tasks and Deprioritising SAE Research (GDM Mech Interp Team Progress Update #2)” by Neel Nanda, lewis smith, Senthooran Rajamanoharan, Arthur Conmy, Callum McDougall, Tom Lieberum, János Kramár, Rohin Shah
Audio note: this article contains 31 uses of latex notation, so the narration may be difficult to follow. There's a link to the original text in th…
1 year, 5 months ago
[Linkpost] “Playing in the Creek” by Hastings
This is a link post. When I was a really small kid, one of my favorite activities was to try and dam up the creek in my backyard. I would carefully m…
1 year, 5 months ago
“Thoughts on AI 2027” by Max Harms
This is part of the MIRI Single Author Series. Pieces in this series represent the beliefs and opinions of their named authors, and do not claim to …
1 year, 5 months ago
“Short Timelines don’t Devalue Long Horizon Research” by Vladimir_Nesov
Short AI takeoff timelines seem to leave no time for some lines of alignment research to become impactful. But any research rebalances the mix of cu…
1 year, 5 months ago
“Alignment Faking Revisited: Improved Classifiers and Open Source Extensions” by John Hughes, abhayesian, Akbir Khan, Fabien Roger
In this post, we present a replication and extension of an alignment faking model organism:
Replication: We replicate the alignment faking (AF) pap…
1 year, 5 months ago