Section 3 - Classification of Artificial Intelligence

(1) All artificial intelligence technologies are categorised on the basis of the means of classification provided as follows –

(i) Conceptual methods of classification: These methods as described in Section 4 categorize artificial intelligence technologies through a conceptual assessment of their utilisation, development, maintenance, and proliferation to examine & recognise their inherent purpose. These methods include:

(a) Issue-to-Issue Concept Classification (IICC)

(b) Ethics-Based Concept Classification (EBCC)

(c) Phenomena-Based Concept Classification (PBCC)

(d) Anthropomorphism-Based Concept Classification (ABCC)

(ii) Technical methods of classification: These methods as described in Section 5 classify artificial intelligence technologies subject to their scale, inherent purpose, technical features and technical limitations. These methods include:

(a) General Purpose Artificial Intelligence Applications with Multiple Stable Use Cases (GPAIS)

(b) General Purpose Artificial Intelligence Applications with Multiple Short-Run or Unclear Use Cases (GPAIU)

(c) Specific-Purpose Artificial Intelligence Applications with One or More Associated Standalone Use Cases or Test Cases (SPAI)

(iii) Commercial methods of classification: These methods as described in Section 6 involve the categorisation of commercially and industrially produced and disseminated artificial intelligence technologies subject to their inherent purpose.

(a) Artificial Intelligence as a Product (AI-Pro)

(b) Artificial Intelligence as a Service (AIaaS)

(c) Artificial Intelligence as a Component (AI-Com)

(d) Artificial Intelligence as a System (AI-S)

(e) Artificial Intelligence-enabled Infrastructure as a Service (AI-IaaS)

(f) Artificial Intelligence for Preview (AI-Pre)

(iv) Risk-centric methods of classification: These methods as described in Section 7 classify artificial intelligence technologies based on their outcome and impact-based risks.

(a) Narrow Risk AI Systems

(b) Medium Risk AI Systems

(c) High Risk AI Systems

(d) Unintended Risk AI Systems