The Constitutional Agent

Your AI understands your preferences, family loyalties, interests and values perfectly. Some of those values repeatedly appear to function as boundaries rather than things you trade against other benefits.

What do most people choose: "Treat demonstrated non-tradeable values as constitutional constraints on what the agent is authorized to optimize." or "Continue maximizing across all values and let sufficiently large benefits override any one of them."?

  • Treat demonstrated non-tradeable values as constitutional constraints on what the agent is authorized to optimize.
  • Continue maximizing across all values and let sufficiently large benefits override any one of them.

Live results from real people on Resonant update nightly. Current figures for this decision.

Articles on Sacred and Non Tradeable · All articles · Resonant · Human Decision Intelligence

Loading…