Hi, thank you for visiting my website!
I care about Burkean Longtermism, which has motivated me to work on preventing existential risks from misaligned AI. Personally, I think the most effective way to contribute to this cause is by working on AI control, which is what I (mostly) concern myself with.
Aside from this, I'm also interested in Transformer interpretability.
This site is pretty bare-bones right now, but I hope to put out a more permanent site soon!
I aspire to
Crocker's Rules, please feel free to be maximally clear and direct in your communication!
Most usernames: @kaushikreddyxyz