> ## Content Index
> Fetch the complete content index at: https://www.aigl.blog/llms.txt
> Use this file to discover other available public pages before exploring further.

# AI, Data Governance and Privacy: Synergies and Areas of International Co-operation
- URL: https://www.aigl.blog/ai-data-governance-and-privacy-synergies-and-areas-of-international-co-operation/
- Published: 2026-09-19T09:06:16.000Z
- Updated: 2026-09-19T09:06:16.000Z
- Description: OECD Artificial Intelligence Paper No. 22 mapping the OECD Privacy Guidelines to the OECD AI Principles, taking stock of privacy enforcement authorities' guidance and enforcement actions on AI, and identifying areas for international co-operation.
- Author: Jakub Szarmach
- Tags: Report, AI Governance, Privacy & Data Protection, Data Governance, AI Policy & Regulation, Responsible AI, #aigl-library

## ⚡ Quick Summary

Published by the OECD in its Artificial Intelligence Papers series (June 2024, No. 22), this report examines the intersection of AI, data governance and privacy and sets out areas for international co-operation. It was approved and declassified by the OECD Digital Policy Committee on 20 June 2024 and prepared by the OECD Secretariat with the OECD.AI Expert Group on AI, Data and Privacy.

The report argues that the AI and privacy policy communities address shared challenges largely in silos, producing divergent terminology, conflicting or duplicative requirements and uncertainty about the actual reach of privacy law, and that generative AI has made this more urgent. It maps the principles of the 1980 OECD Privacy Guidelines, revised in 2013, onto the five values-based principles of the 2019 OECD Recommendation on AI, updated in 2024, comparing how each community reads fairness, transparency and explainability.

It takes stock of initiatives by Privacy Enforcement Authorities, including the G7 Data Protection and Privacy Authorities statement, the Global Privacy Assembly resolution on generative AI systems and enforcement actions concerning ChatGPT and facial recognition, and points to Privacy Enhancing Technologies as an area for collaboration. It concludes that AI's innovative, technological and regulatory developments are mainly compatible with, and can reinforce, privacy and personal data protection rules.

## 🧩 What’s Covered

- **Framing and definitions:** the Introduction defines the "AI and privacy communities" and "AI actors", generative AI, and privacy and data protection, and sets out the conclusion that AI developments are mainly compatible with privacy rules.
- **Generative AI as a catalyst:** Chapter 1 covers opportunities, including Privacy Enhancing Technologies such as homomorphic encryption, trusted execution environments, federated learning, differential privacy, synthetic data and machine unlearning, alongside risks such as inference of personal attributes, failing de-identification, model memorisation and obstacles to rectification and erasure.
- **Privacy Enforcement Authorities step in:** three initiatives are detailed — the G7 Roundtable statement of June 2023, the GPA International Enforcement Co-operation Working Group statement on web scraping of 24 August 2023, and the GPA Resolution on Generative Artificial Intelligence Systems of 20 October 2023 with its nine principles.
- **Mapping principles:** Chapter 2 presents Table 1 (the five values-based AI principles and five recommendations to governments), Table 2 (similarities and areas of coordination by principle) and Table 3 (key concepts with different meanings), noting that privacy considerations are mapped onto the AI Principles.
- **Principle-by-principle analysis:** each value-based principle is examined — inclusive growth and well-being; rule of law, human rights, fairness and privacy, including sources of bias, data minimisation, purpose specification and the CNIL AI how-to sheets; transparency and explainability, including NIST's four principles and AEPD guidance; robustness, security and safety, including adversarial machine learning and red teaming; and accountability through privacy management programmes and the AI system lifecycle.
- **Terminology comparisons:** fairness, transparency and explainability, and privacy and data protection are contrasted between the two communities, including the French terms "équité" and "loyauté".
- **National and regional developments:** Chapter 3 lists guidance from Canadian privacy regulators, the CNIL, the AEPD, Türkiye's KVKK, the UK ICO, the US FTC and Singapore's PDPC, and enforcement actions concerning ChatGPT, Clearview AI, Rite Aid and Amazon.
- **Conclusion:** the report sets out the role of the OECD.AI Expert Group on AI, Data and Privacy and possible future work on common guidance, sector-specific issues and whether the AI and Privacy Recommendations need updating.

## 💡 Why it matters?

The report gives policy, compliance and oversight teams a structured way to see where AI policy and privacy law overlap, where terminology diverges, and where requirements risk duplication. Its mapping of the OECD Privacy Guidelines to the OECD AI Principles, and its inventory of PEA guidance and enforcement, help practitioners judge how existing privacy obligations already reach AI systems and where coordination with AI-specific frameworks — such as the EU AI Act, NIST work and Human Rights Impact Assessments — is needed. It is a reference for framing interoperability rather than a compliance checklist.

## ❓ What’s Missing

The national and regional overview is explicitly current only "as early 2024" and does not claim to be exhaustive, so later guidance and enforcement are absent. The mapping covers only the five values-based AI principles; analysis of the five recommendations to governments is deferred to possible subsequent work. Several substantive questions are left open, including whether individuals can have data inside a model corrected or deleted and whether a whole model should be deleted, and how data minimisation applies to training data. The report offers no sector-specific operational steps.

## 👥 Best For

Privacy regulators, data protection officers and AI policy teams seeking a shared vocabulary; compliance leads assessing how existing privacy law already applies to AI systems and where AI-specific rules intersect; and analysts tracking international cooperation between privacy and AI governance communities.

## 📄 Source Details

*AI, Data Governance and Privacy: Synergies and Areas of International Co-operation*, OECD Artificial Intelligence Papers No. 22, published by the OECD, June 2024, 55 pages, reference code DSTI/CDEP/AIGO/DGP(2023)1/FINAL, © OECD 2024\. The document carries English and French abstracts and executive summaries; the main language is English. The input was a text extraction of all 55 pages. No URL for the document itself is printed; only the OECD terms and conditions address and links to other works appear.