DeepMind
Risk assessment 29%
50%
Evals: domains, quality, elicitation
1%
Evals: accountability
0%
Adversarial evaluation for alignment
10%
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
50%Click to show details/rubric
Evals: accountability
1%Click to show details/rubric
Adversarial evaluation for alignment
0%Click to show details/rubric
Model organisms
10%Click to show details/rubric