Human-looking robots are a bad idea
epistemic status: opinionated view on the dangers of robots that look like humansIt's not a…
Category: Opinion
How Go Players Disempower Themselves to AI
Written as part of the MATS 9.1 extension program, mentored by Richard Ngo.From March 9th…
Early-stage empirical work on “spillway motivations”
Previously, we proposed spillway motivations as a way to mitigate misalignment induced via training a model…
Exploration Hacking: Can LLMs Learn to Resist RL Training?
We empirically investigate exploration hacking (EH) — where models strategically alter their exploration to resist…
Conditional misalignment: Mitigations can hide EM behind contextual cues
This is the abstract, introduction, and discussion of our new paper. We study three popular mitigations…
![Ambitious Mech Interp w/ Tensor-transformers on toy languages [Project Proposal]](https://res.cloudinary.com/lesswrong-2-0/image/upload/v1777662479/lexical_client_uploads/wdemj6iasqsre619ndeu.png)
Ambitious Mech Interp w/ Tensor-transformers on toy languages [Project Proposal]
This is my project proposal for Pivotal. Apply as a mentee by May 3rdThe field…
Your four-dimensional body
IntroductionYou are a four-dimensional being.And it’s fitting, because you live in a four-dimensional world. Three…
Housing Roundup #14: You Can’t Build That
Why can’t you build it? Because you aren’t allowed to build it. Not in the…
What do Russian olympiad winners think of HPMOR? Our data
I gave Claude access to our stats and asked it to generate a page presenting…
Reflections on InkHaven
If you’ve been wondering why I’m suddenly blogging every day… well, it’s about to stop!…