26 ms·Arguments about Highly Reliable Agent Designs as a Useful Path to AI Safety1 points by CapitalistCartr 5y ago