Andrew runs Arcadia Impact's alignment research team. He has a PhD in theoretical computer science from Aarhus University and previously worked on data poisoning.
Posts
Labs’ current alignment plans seem terrible
Daniel Tan, Andrew Draganov, Nicole Xing
Alignment Midtraining Cracks Under Pressure
Jonathan Bostock, Sid Baines, Daniel Tan, Andrew Draganov, Angel Martinez, David Africa
Automated alignment runs are hard to study!
Alejandro Aristizabal, Andrew Draganov, Aleksandr Bowkis, Cameron Holmes
Your Model Organisms Might Be Fried
Daniel Tan, Jonathan Bostock, Andrew Draganov, Angel Martinez, Sid Baines, David Africa
Learnings from starting an AI safety research team
Andrew Draganov, Erin Robertson