Every AI Policy Ever · open in the interactive register

Published (non-binding, current)Non-binding · Guidelines / framework

Ethical Artificial Intelligence Framework

Record HK-01

Issued by
Digital Policy Office, HKSAR Government (formerly Office of the Government Chief Information Officer)
Jurisdiction
Hong Kong SAR (China)
Level · region
Sub-national · East Asia
Country
Hong Kong SAR (China)
Published / adopted
2021
Last updated
2025-12
Language
English, Chinese

Summary

Framework for government bureaux and departments, also offered in adapted form to other organisations, on building ethics into AI and big-data projects. It sets 12 Ethical AI Principles (including transparency and explainability, reliability, robustness and security, human oversight, safety and accountability) and an AI Application Impact Assessment. High-risk projects need IT Board or CIO review, and assessments are repeated at four life-cycle stages.

Safety topics

public-sector userisk-based classificationhuman oversightaccountability & liabilitytransparency & labelling

Status notes

First introduced in 2021 and updated in 2023, according to the Secretary for Innovation, Technology and Industry (30 Oct 2025). Current text is version 2.0 (December 2025), which updated Sections 1 to 5 and Appendices A to D and removed Appendix E. Earlier versions: 1.1 (June 2023), 1.2 (July 2023), 1.3 (August 2023) and 1.4 (July 2024, which replaced 'Office of the Government Chief Information Officer' with 'Digital Policy Office'). The framework was first developed for government bureaux and departments; this version is adapted for general reference by other organisations. Where the Chinese and English versions differ, the English version prevails.

Sources

Open the official source https://www.digitalpolicy.gov.hk/en/our_work/data_governance/policies_standards/ethical_ai_framework/

Secondary source: https://www.digitalpolicy.gov.hk/en/our_work/data_governance/policies_standards/ethical_ai_framework/doc/Ethical_AI_Framework_en.pdf

Local copy

Archived. Last attempt 2026-10-18 06:48:50 UTC

Official source (third-party copy, not published) archive/HK-01/2026-10-11/official.html
Retrieved 2026-10-11 14:04:28 UTC
SHA-256 f3fe4ff8caf13d528455dbf02009f6adfc21f85ed6852c1b2f330e8ca93ad2a0
Linked document (PDF) (third-party copy, not published) archive/HK-01/2026-10-11/document-1.pdf
Retrieved 2026-10-11 14:04:36 UTC
SHA-256 8cf14d321262aae177a739f8f54bd5d7373ffad58a004224654fd05d2adcacd2
Linked document (PDF) (third-party copy, not published) archive/HK-01/2026-10-11/document-2.pdf
Retrieved 2026-10-11 14:04:44 UTC
SHA-256 a0ec7fa02e162e73f7f10939896815425e49e80ef29cacf770bb8de370eecf0b
Secondary source (third-party copy, not published) archive/HK-01/2026-10-11/secondary.pdf
Retrieved 2026-10-11 14:04:49 UTC
SHA-256 a0ec7fa02e162e73f7f10939896815425e49e80ef29cacf770bb8de370eecf0b

Verification

B
Another official source, such as an announcement by the issuer, confirms the title, date and status. checked 2026-10-18 · alternative-official-url-fetched
Refresh of 18 Oct 2026 under the rule that level A needs the instrument's own text or an official register. Re-read the framework PDF, version 2.0 dated December 2025 (digitalpolicy.gov.hk, stored 11 Oct 2026, unchanged on 18 Oct 2026), and the Digital Policy Office's page 'AI and Data Ethics' that presents it and links the PDF (same dates). The PDF confirms the title, the issuer, the version and its date, the amendment history from version 1.1 (June 2023) and the content of the summary. Neither the PDF nor the page dates the first version: date_published, the year 2021, rests on a speech by the Secretary for Innovation, Technology and Industry of 30 Oct 2025 on the bureau's site (itib.gov.hk, stored again in this run on 18 Oct 2026), which says the framework was 'first introduced in 2021' and updated in 2023. A speech is an announcement, so the date counts as alt-official and the record is level B until the framework's first version or an official record of its issue is stored.
Second check: Independent check by Claude Code (claude-sonnet-5-5), 2026-10-18: 8 field(s) supported.

History

  1. 2026-10-18verification.level: A → B — level computed from the evidence by tools/pipeline.py2026-10-18-refresh-level-rule-1 · Claude Code (claude-opus-5-5)
  2. 2026-10-18notes: Current text is version 2.0 (December 2025). Earlier versions: 1.1 (June 2023), → First introduced in 2021 and updated in 2023, according to the Secretary for Inn — the year of first introduction and the 2023 update added from the speech; the remark that the release date of version 1.0 is left null concerns the record and was removed; version 2.0's changes, the adaptation for general use and the English-prevails clause added from the PDF2026-10-18-recheck-east-asia-1 · Claude Code (claude-opus-5-5)
  3. 2026-10-18date_published: – → 2021 — was null; the Secretary for Innovation, Technology and Industry's speech of 30 Oct 2025 says the framework was 'first introduced in 2021', so the year is recorded at that precision2026-10-18-recheck-east-asia-1 · Claude Code (claude-opus-5-5)

Related records

Hong Kong Generative AI Guideline Hong Kong SAR (China) · Published (non-binding, current)

Citation

Digital Policy Office, HKSAR Government (formerly Office of the Government Chief Information Officer). Ethical Artificial Intelligence Framework. 2021. https://www.digitalpolicy.gov.hk/en/our_work/data_governance/policies_standards/ethical_ai_framework/. Status: Published (non-binding, current); verified 2026-10-18 (level B). Local copy archive/HK-01/2026-10-11/official.html retrieved 2026-10-11 14:04:28 UTC, SHA-256 f3fe4ff8caf13d528455dbf02009f6adfc21f85ed6852c1b2f330e8ca93ad2a0. Every AI Policy Ever, version 1.18 (snapshot 2026-10-18), record HK-01.

Improve this record

Know of an amendment, a status change or an error? Report news or an error on GitHub; the form opens with this record filled in.