OK interesting. To put it mildly, that feels...speculative.
I would think it makes more sense to start with proven, immediate things & then expand outwards from there. Global Health feels like an obvious #1 to me, whereas AI-alignment feels like a #8.
Putting a probability on AI-alignment issues is a complete shot in the dark. Nobody knows. Whereas there is zero dispute about the effectiveness of immunizations.
Am I missing something? This logic feels painfully obvious to me. What am I missing?
OK interesting. To put it mildly, that feels...speculative.
I would think it makes more sense to start with proven, immediate things & then expand outwards from there. Global Health feels like an obvious #1 to me, whereas AI-alignment feels like a #8.
Putting a probability on AI-alignment issues is a complete shot in the dark. Nobody knows. Whereas there is zero dispute about the effectiveness of immunizations.
Am I missing something? This logic feels painfully obvious to me. What am I missing?