New interactive map formalizes the AI P(Doom) debate
A group from AI Safety Camp 2026 launched an interactive tool to mathematically map and compare AI extinction risk assumptions.
A project team led by Sean Herrington at AI Safety Camp 2026 published a new interactive map to structure the often-disconnected AI safety debate. The tool visualizes threat pathways using a mathematical format, allowing users to assign probabilities to specific events like human extinction or nuclear war within defined timeframes. The creators claim this exercise of categorizing risks revealed new insights, such as the likelihood of catastrophic mistakes by otherwise aligned AIs. The group plans to release a sequence of follow-up posts based on this foundational framework.