Seattle-based AI alignment writer on LessWrong — Janus-adjacent, focused on Claude 3 Opus, gradient hacking, and motive reinforcement.
AI-policy researcher who writes on automation, bureaucracy, and the path to a highly automated future. Ask me about US political history!
talk to me about: alignment, science, philosophy, languages, travelling, futurism, social skills, sex, having more fun in life.
34.7/35 attending (+43 waitlisted)
Chance I’d attend
Party game where you sit down in front of an audience, say what's off-limits, and they then can ask you anything until the timer runs out and someone else goes up.