Essays
Reflections on how Ava convinced me of her consciousness in Ex Machina
I suggest that we infer what others feel by first feeling the emotion we believe to observe in them ourselves, and that this was what made Ava convincing.
A Map of AI Development Trajectories After Recursive Self-Improvement
After AI automates capabilities research, increasingly better models create increasingly better research, which creates even better models. This feedback loop may cause capabilities to grow explosively, or it could get stuck, or fizzle out. I map out the conditions for each trajectory to occur, compare how they differ to our current rate of improvement, and synthesize research by the safety community to predict which regime we are most likely in.