Anthropic
Risk assessment 49%
60%
Evals: domains, quality, elicitation
15%
Evals: accountability
50%
Adversarial evaluation for alignment
75%
Model organisms
AI companies should do model evals and uplift experiments to determine whether models have dangerous capabilities or how close they are. They should also prepare to check whether models will act well in high-stakes situations.
Evals: domains, quality, elicitation
60%Click to show details/rubric
Evals: accountability
15%Click to show details/rubric
Adversarial evaluation for alignment
50%Click to show details/rubric
Model organisms
75%Click to show details/rubric