Principle 2: Protection

Throughout the entire lifecycle of AI systems, protective and security measures must be implemented to guarantee protection against attacks and unwanted damage.

See also: