x

AI ALIGNMENT FORUM

AF

Caleb Biddulph — AI Alignment Forum

Caleb Biddulph

Top postsTop post

Caleb Biddulph

Message

Currently doing independent AI safety research and blogging!

In reverse date order, I've been a:

Fellow in Astra Fellowship 2.0 with Redwood Research, mentored by Adam Kaufman
- I wrote about my work in Untrusted advice for AI control: Short, strong advice significantly uplifts weak LLMs
MATS 8.1 scholar, mentored

...

3451

Ω

37

19

329

1

5y

Caleb Biddulph

Currently doing independent AI safety research and blogging!

In reverse date order, I've been a:

Fellow in Astra Fellowship 2.0 with Redwood Research, mentored by Adam Kaufman
- I wrote about my work in Untrusted advice for AI control: Short, strong advice significantly uplifts weak LLMs
MATS 8.1 scholar, mentored

...

5 ways to improve CoT faithfulness

In the case for CoT unfaithfulness is overstated, @nostalgebraist pointed out that reading the chain-of-thought (CoT) reasoning of models is neglected as an interpretability technique. I completely agree, but I also believe that with our current methods, there's a significant risk of being misled by the CoTs of models like...

Oct 5, 2024•46

Preparing for AI-assisted alignment research: we need data!

Thanks to Neel Nanda, Madeleine Chang, Michael Chen, Oliver Zhang, and Aidan O’Gara for their early feedback on this post. Summary: Alignment researchers could record the intermediate outputs of their work in anticipation of using the data to train future AI models. These models could eventually accelerate alignment research by...

Jan 17, 2023•31