flâneur

How Can AI Labs Incorporate Risks From AI Accelerating AI Progress Into Their Responsible Scaling Policies?

forethought.org · 3,120 words · saved by 1 readers

I describe a threat model by which AI R&D capabilities could cause harm, a specific capability threshold at which this risk becomes unacceptable, early warning signs for detecting that threshold, and the protective measures needed to continue development safely past that threshold. I recommend that labs start measuring for the warning signs today. If they observe them, they should pause AI development unless they have implemented the protective measures.

How Can AI Labs Incorporate Risks From AI Accelerating AI Progress Into Their Responsible Scaling Policies? Tom Davidson Citations Cite Citations PDF Contact 24th March 2025 How Can AI Labs Incorporate Risks From AI Accelerating AI Progress Into Their Responsible Scaling Policies? Summary Threat model Early warning signs AI has already doubled the pace of software progress AI can autonomously perform many AI R&D tasks Protective measures [AI Narration] How Can AI Labs Incorporate Risks From AI Accelerating AI Progress Into Their Responsible Scaling Policies? Playback speed 0.5 × 0.75 × 1 × 1.2

saved by

related reading