OAISI x EA: Priorities in AI Safety

​Join EA Oxford and the Oxford AI Safety Initiative for a panel discussion on priorities for ensuring the development of transformative AI goes well for humanity. We'll host an interactive discussion on some key themes and debates with AI safety researchers and strategists in OAISI's network. We encourage audience participation, and hope to continue the conversation over pizza afterwards!

​​(i) which problems posed by transformative AI seem most pressing? What normative and empirical assumptions does this depend on?

​​(ii) which considerations seem particularly important for high-level strategy around AGI? What seems necessary/important for AGI to go well?

​​(iii) what kinds of directions within AI safety should more impact-motivated people be pursuing?