• Scenario 1: Alignment is much easier than expected
• Scenario 2: The combination of many technical and governance strategies works
• Scenario 3: Accident and regulation (warning shot + moratorium)
• Scenario 4: Alignment by chance (we get lucky and the first TAI is fairly intent aligned)
• Scenario 5: US-China-driven global cooperation (moratorium, IAEA for AI)
• Scenario 6: Apollo Project for AI (well-resourced huge safe AI project)
• [notes]
  • Stories of how AI goes well. Not action-oriented but describes potentially good paths.
    • It worries me a little that no path seems like an excellent option