Conversations with AI: AI Governance

This article is about AI Governance while chatting with one of most famous AI instruments “ChatGPT o1-Preview”. Like my previous articles on similar subjects, this is a chat between me and AI. For this article I chose Chatgpt-01 Preview because it got updated lately and it scored high on reasoning.

In the discussions below my questions or requirements from the new ChatGPT o1 Preview are under my name and ChatGPT-o1 Preview answered. In the first prompt you will notice that I asked ChatGPT-o1 Preview to set creativity temperature to 0. This means it should not create any stories but provide information only. This decreases AI hallucinations and forces it to adhere to facts.

Walid:

Today we will chat about "AI Governance", please set creativity temperature to 0 and provide reference for any statistics or other data. Are you familiar with ISO, OECD, UN and EU requirements for AI governance?

ChatGPT said:

Thought for 11 seconds

Yes, I am familiar with the requirements and guidelines for AI governance established by the ISO, OECD, UN, and the EU.

• ISO: The International Organization for Standardization (ISO) is developing standards for AI through its subcommittee ISO/IEC JTC 1/SC 42. These standards address various aspects of AI, including terminology, trustworthiness, and ethical concerns. Source: ISO/IEC JTC 1/SC 42 Artificial Intelligence
• OECD: The Organisation for Economic Co-operation and Development (OECD) adopted the OECD AI Principles in May 2019(*1). These principles promote AI that is innovative, trustworthy, and respects human rights and democratic values. Source: OECD AI Principles
• UN: The United Nations, through UNESCO, adopted the "Recommendation on the Ethics of Artificial Intelligence" in November 2021. This framework provides ethical guidelines for the development and deployment of AI technologies. Source: UNESCO Recommendation on the Ethics of AI
• EU: The European Union is in the process of establishing the Artificial Intelligence Act, a regulatory framework aiming to ensure that AI systems are safe, transparent, and respect fundamental rights. Source: Proposal for a Regulation laying down harmonised rules on artificial intelligence (Artificial Intelligence Act)

Please let me know if you need more detailed information on any of these guidelines.

o1

Walid:

Please tell me how do companies govern AI? Please provide examples and comparisons on how to utilize international standards and requirements to enhance AI governance.

ChatGPT said:

Thought for 10 seconds

How Companies Govern AI and Utilize International Standards to Enhance AI Governance

Companies govern Artificial Intelligence (AI) by establishing frameworks, policies, and practices that ensure the ethical, responsible, and lawful development and deployment of AI technologies. They utilize international standards and requirements to align their AI governance with global best practices, enhance trust, and comply with regulatory expectations.

1. Internal AI Governance Structures

a. Establishing Ethical Guidelines and Principles

Microsoft: Developed Responsible AI Principles focusing on fairness, reliability, safety, privacy, inclusiveness, transparency, and accountability. These principles guide all AI development within the company.

Reference: Microsoft AI Principles

Google: Adopted a set of AI Principles that emphasize socially beneficial AI, avoiding creating or reinforcing bias, and ensuring safety and accountability.

Reference: Google AI Principles

IBM: Implemented Ethics for AI guidelines that prioritize transparency, explainability, fairness, and robustness in AI systems.

Reference: IBM AI Ethics

b. Forming Oversight Committees and Boards

Microsoft's Aether Committee: Advises on issues, policies, and processes related to responsible AI. It includes experts from various disciplines to oversee AI initiatives.

Google's AI Ethics Council: Although faced challenges and was dissolved, Google continues to have internal review processes to assess AI projects against ethical standards.

IBM's AI Ethics Board: Oversees the ethical implications of AI technologies and ensures compliance with internal and external guidelines.

c. Implementing Training and Awareness Programs

Companies conduct regular training for employees to ensure understanding and adherence to AI governance policies.

2. Utilizing International Standards and Requirements

Companies enhance their AI governance by aligning with international standards and regulatory frameworks, ensuring their practices meet global expectations.

a. Adoption of ISO Standards

ISO/IEC JTC 1/SC 42: Companies adopt standards developed by the International Organization for Standardization (ISO) for AI, which cover AI concepts, terminology, and trustworthiness.

Example: Siemens incorporates ISO standards in their AI development processes to ensure quality and reliability.

Reference: Siemens and AI Standards

b. Alignment with OECD AI Principles

OECD AI Principles: Encourage AI that is innovative, trustworthy, and respects human rights.

Example: SAP aligns its AI ethics guidelines with OECD principles to promote transparency and accountability in its AI solutions.

Reference: SAP Guiding Principles for AI

c. Compliance with EU Regulations

EU's Artificial Intelligence Act: Sets harmonized rules on AI to ensure safety and fundamental rights protection.

Example: Airbus adjusts its AI applications in compliance with the upcoming EU AI regulations, focusing on risk management and compliance.

Reference: Airbus and AI Compliance

d. Following UN and UNESCO Guidelines

UNESCO's Recommendation on the Ethics of AI: Provides a global normative framework for ethical AI.

Example: Telefonica incorporates UNESCO's ethical guidelines into its AI governance model to ensure respect for human rights and ethical principles.

Reference: Telefonica's AI Principles

3. Comparisons and Implementation Strategies

a. Holistic Integration vs. Specific Compliance

Holistic Integration: Companies like Microsoft and IBM integrate international standards into every aspect of their AI governance, creating comprehensive policies that exceed basic compliance.

Specific Compliance: Companies such as Airbus focus on complying with specific regulations like the EU AI Act, tailoring their AI governance to meet these requirements.

b. Proactive Ethical Leadership vs. Reactive Compliance

Proactive Ethical Leadership: Google and SAP proactively develop ethical guidelines aligned with international principles, aiming to lead in responsible AI.

Reactive Compliance: Some companies may primarily react to regulatory changes, updating their governance frameworks in response to new laws to ensure compliance.

c. Global Standards Adoption vs. Regional Focus

Global Standards Adoption: Multinational corporations often adopt global standards like those from ISO and OECD to maintain consistency across all operations.

Regional Focus: Companies operating primarily within specific regions may focus on local regulations, such as the EU's AI Act or China's AI regulations.

4. Benefits of Utilizing International Standards

Consistency: Ensures uniform AI governance practices across different jurisdictions.

Trust and Reputation: Enhances stakeholder trust by demonstrating a commitment to global best practices.

Risk Management: Reduces legal and ethical risks by adhering to recognized standards.

Innovation Facilitation: Encourages responsible innovation by providing clear guidelines.

5. Challenges and Considerations

Regulatory Complexity: Navigating and reconciling different international standards can be complex.

Cultural Differences: Ethical perceptions of AI can vary across cultures, affecting governance approaches.

Resource Allocation: Implementing comprehensive AI governance aligned with international standards requires significant resources.

Conclusion

Companies govern AI by establishing internal frameworks that are often enhanced through the adoption of international standards and requirements. By aligning with global principles from organizations like ISO, OECD, UN, and regulatory bodies like the EU, companies can enhance their AI governance, ensuring responsible, ethical, and compliant AI development and deployment. This alignment not only helps in managing risks but also builds trust with stakeholders and promotes sustainable innovation.

Please let me know if you need further details on any of these points.

Notes by Walid:

*1 – SCOPING THE OECD AI PRINCIPLES DELIBERATIONS OF THE EXPERT GROUP ON ARTIFICIAL INTELLIGENCE AT THE OECD (AIGO), OECD DIGITAL ECONOMY PAPERS November 2019 No. 291

I always advise my clients that if you have different standards to apply in different operations, being EU or OECD requirements or the like, always apply the stricter requirements if possible. This will protect the company from being sued in any of the different jurisdictions.