On 20 September 2024, the Malaysian Government, through its Ministry of Science, Technology and Innovation (“MOSTI”), launched the National Guidelines on AI Governance & Ethics (“AIGE”). The launch of the AIGE aims to support the implementation of the Malaysian National Artificial Intelligence Roadmap 2021-2025 (“AI-RMAP”), which seeks to establish frameworks for AI integration across various economic sectors through policy initiatives, while the Government develops laws to regulate the use of AI.

The 7 AI Principles

The AIGE outlines the following seven fundamental AI principles for the practice of responsible AI:

1. Fairness

AI systems should be designed to avoid bias or discrimination against the target users.

2. Reliability, Safety and Control

AI systems should operate reliably and consistently under both normal and unexpected conditions to ensure proper access, control, and protection in critical situations.

3. Privacy and Security

AI systems should be safe, secure, and perform as intended, resisting compromise by unauthorised parties.

4. Inclusiveness

AI must be inclusive for all stakeholders to prevent unequal access that could exacerbate social divides.

5. Transparency

AI algorithms should be transparent to enable stakeholders to understand both the technical processes of an AI system and related human decisions, and evaluate risks and address any issues that may arise. This includes:

a. Full disclosure regarding the use of AI in decision-making and its intended purpose;

b. Information on training data and any historical or social biases;

c. Procedures for data quality verification;

d. Maintenance and assessment protocols; and

e. Mechanisms for challenging AI decisions

6. Accountability

Developers, owners, and operators of AI should be accountable for ensuring the system’s performance and compliance with AI Acts, governance, and ethical principles. The four elements of system purpose, technology capability, quality and reliability, and consideration of sensitive users, must be addressed to avoid consequential harm.

7. The Pursuit of Human Benefit and Happiness

AI systems should respect human-centred values, enhance quality of life, and promote happiness, while allowing human intervention so as to preserve autonomy.

The Stakeholders

The AIGE focuses on three key categories of stakeholders that are essential to establishing the fundamental building blocks of an AI system by applying the seven AI principles to these groups.

1. AI End Users

Essentially, AI end users are individuals or organisations who use AI products, which are amongst others, AI-powered virtual assistants, smart home devices, or smartphone applications. Therefore, different user groups engage with AI products in various ways.

For example, industries such as the government, manufacturing, and banking employ AI for security, quality control, fraud detection, and content creation.

However, many users may not realise they are adopting AI technologies and may not be aware of the associated benefit or risks. These guidelines aim to raise awareness of responsible AI use and protect end users by incorporating principles such as:

  • The right to be informed about data collection;
  • The right to object to unconsented data usage;
  • The right to request for personal data deletion; and
  • The right to collective redress.

In this regard, the AIGE provides a framework to inform the end users of their rights and responsibilities, empowering them to make informed decisions regarding the benefits and risks of using AI technologies and applications.

2. Government, Agencies, Organisations, and Institutions

Government officials, policymakers, and regulatory agencies are responsible for formulating and implementing policies and regulations related to AI, ensuring ethical and responsible AI practices across different sectors. The AIGE aims to provide a framework for these bodies to develop policies and regulations that promote the development and deployment of AI, while preventing the misuse of AI technology.

The presence of regulations, standards, and ethical guidelines governing AI is likely to increase consumer trust and acceptance of these technologies. The recent enactment of the Cybersecurity Act 2024 and the recent passing of the PDPA (Amendment) Bill 2024 represents initial steps towards this goal.

3. Developers, Designers, Technology Providers, and Suppliers

The AIGE highlights best practices, ethical considerations, and technical standards for responsible AI development and deployment. It recommends that stakeholders adopt the following approach to ensure that AI technologies align with societal values and contribute positively to individuals and communities:

(i) Data Collection

Ensuring the data used to train AI models is diverse, representative, and free from biases. This includes identifying potential sources of bias in the data and taking steps to mitigate them.

(ii) Transparency

Providing clear explanations and justifications for decisions made by AI algorithms, making the decision-making process understandable to users.

(iii) Ethical Considerations

Incorporating ethical considerations into the design and development of AI algorithms, proactively identifying and addressing potential ethical challenges.

(iv) Validation

Conducting rigorous testing and validation to ensure the AI algorithm performs as accurately as intended and is reliable.

(v) User Feedback

Incorporating user feedback into the algorithm development process by, amongst others, actively seeking input from users and incorporating their perspectives and needs into the design of the AI system.

(vii) Ongoing Monitoring

Continuously monitoring the AI algorithms’ performance and impact to identify and address any biases, risks, or unintended consequences.

While the AIGE offers general guidance, it also encourages initiative from sectoral industry stakeholders to carefully evaluate and adapt the seven core principles to craft industry-specific guidelines for responsible AI development and deployment.

Conclusion – The Significance of the AIGE

The AIGE marks a significant step toward responsible AI practices in Malaysia, aligning with the government’s goal of enhancing AI development and deployment. It reflects the Malaysian Government’s commitment to human rights, privacy, and societal well-being, aiming to meet international standards for sustainable AI development and corporate responsibility. The AIGE will serve as the foundation for future legislation in relation to the governance and ethical standards of AI technology in Malaysia. Therefore, it would be prudent for stakeholders to align their practices with the AIGE now to ensure a smoother transition when these laws are implemented. 

If you have any queries, please contact Associates Poh Jonn Sen ([email protected]) and Bryson Chee ([email protected]), or their team Partner, G. Vijay Kumar (vkg@lh-ag.com).