Announcing the Safe Pareto Improvements (SPI) Fundamentals Program

According to the Center for Long-Term Risk, Safe Pareto Improvements are a negotiation framework designed to reduce conflict between AI agents. CLR is launching a Fundamentals Program to help researchers understand this approach to AI safety. The four-week online course runs August third through August twenty-eighth, with five to seven hours of weekly work. Participants will study bargaining problems, conflict mitigation, and real-world applications—including how language models handle SPI safety failures. Applications close July twenty-fourth.

Listen to this story

Hear this and more stories in a personalized audio briefing.

Open The Chonkerton