guidance

High-risk § 3.2.2

The critical digital infrastructure use case

Official source ↗

(193)

According to the approach presented in section 3.2.1.ii above, ‘an AI system intended to be used in the management and operation of critical digital infrastructure’ refers to an AI system intended to be used in an asset, a facility, equipment, a network, or a system or a part of an asset, a facility, equipment, a network or a system which is necessary for the provision of the essential services established in Article 2(8) CER Delegated Regulation. This includes:

  • the provision and operation of internet exchange point service s (providers of internet Exchange Points);
  • the provision of domain name system (DNS) service s, excluding services related to root name servers (DNS service providers);
  • the operation and administration of top-level domain (TLD) name registries (TLD name registries);
  • the provision of cloud computing services (providers of cloud computing services);
  • the provision of data centre services (providers of data centre services);
  • the provision of content delivery networks (providers of content delivery networks);
  • the provision of trust services (trust service providers);
  • the provision of publicly available electronic communications services (providers of electronic communications services);
  • the provision of public electronic communications networks (providers of public electronic communications networks).
(194)

An AI system intended to be used in the management and operation of critical digital infrastructure should therefore be considered as high-risk only if it is intended to be used by an entity that has been identified as a critical entity by a Member State under the CER Directive.

a) Practical example of an AI system falling within the high-risk use case of critical digital infrastructure

b) Practical examples of AI systems falling outside the high-risk use case of critical digital infrastructure