• To light
  • Evals to identify AI risks
  • Using eval results to rally support for a slow down
  • Global cooperation in body ‘AWSAI’ to safely deploying strong AI
  • AI used to create life extension treatments. Schism forms between people who use life extension (the ‘tempered’) and those who don’t.
  • People are happy in small groups with similar interests (cliques)
  • Land value taxes
• Peace through prophecy
  • Governments use prediction markets, liquid democracy and AI research tools to make wiser decisions and capably execute big projects.
    • Positive feedback loop as people copy countries that do this successfully.
  • Flash minor war (a few ships sunk) due to feedback loop in military AI between US and China.
  • Global agreement on AI (delphi accords) to tackle misalignment. Spurred by AI tools helping governments make better decisions, and by pressure after flash war (which acted like a fire alarm event).
• Core central
  • Risks from AI became more apparent, and so funding increased for control, alignment and explainability. A few runaway AI scares encouraged global cooperation, including regulation requiring AI systems to be aligned and explainable. Nations not agreeing were sanctioned.
  • The first AGI develops a ‘Core’ system which countries adopt for governance
    • [I don’t know what core is after trying to read this for a while; I gave up]
• [notes]
  • This was less helpful than I thought it would be