AI Glossary

All 1 A B C D E F G H I J K L M N O P Q R S T U V W Z

C

Capability vs. Safety Evaluation

The distinction between evaluating what a model can do (its capabilities) and evaluating whether its use is safe; both form part of managing systemic risks.

Compliance Architecture for AI Agents

System design in which legal and security requirements, defined purposes of use, risk management and ongoing auditability are built in from the development stage onwards, so that the agent remains compliant with regulation.