Racing to
the Precipice
Computational Model
Original publication by Stuart Armstrong, Nick Bostrom, and Carl Shulman
https://www.fhi.ox.ac.uk/wp-content/uploads/Racing-to-the-precipice-a-model-of-artificial-intelligence-development.pdf
Number of agents
Number of iterations:
Agent behavior:
No information Nash equilibrium
Private information Nash equilibrium
Public information Nash equilibrium
Max capability:
Enmity:
Randomness seed:
Run model
Save as SVG
Save as PNG
View Source
Open in Vega Editor
Simulated disaster probability: 0.610
Analytical disaster probability: 0.600
Randomness seed used: 3808069367042714