The AI Containment Problem | Roman Yampolskiy

Can a superintelligent AI ever truly be contained?* AI safety researcher Roman Yampolskiy argues that sufficiently advanced …

source