Safe and Ethical AI (SEA) Platform Network · Linking Artificial Intelligence Principles (LAIP)

Auditability

Enable interested third parties to probe, understand, and review the behavior of the algorithm through disclosure of information that enables monitoring, checking, or criticism, including through provision of detailed documentation, technically suitable APIs, and permissive terms of use.

Principle: Principles for Accountable Algorithms, Jul 22, 2016 (unconfirmed)

Published by Fairness, Accountability, and Transparency in Machine Learning (FAT/ML)

Related Principles

5. Privacy and Data Governance

AI systems should have proper mechanisms in place to ensure data privacy and protection and maintain and protect the quality and integrity of data throughout their entire lifecycle. Data protocols need to be set up to govern who can access data and when data can be accessed. Data privacy and protection should be respected and upheld during the design, development, and deployment of AI systems. The way data is collected, stored, generated, and deleted throughout the AI system lifecycle must comply with applicable data protection laws, data governance legislation, and ethical principles. Some data protection and privacy laws in ASEAN include Malaysia’s Personal Data Protection Act 2010, the Philippines’ Data Privacy Act of 2012, Singapore’s Personal Data Protection Act 2012, Thailand’s Personal Data Protection Act 2019, Indonesia’s Personal Data Protection Law 2022, and Vietnam’s Personal Data Protection Decree 2023. Organisations should be transparent about their data collection practices, including the types of data collected, how it is used, and who has access to it. Organisations should ensure that necessary consent is obtained from individuals before collecting, using, or disclosing personal data for AI development and deployment, or otherwise have appropriate legal basis to collect, use or disclose personal data without consent. Unnecessary or irrelevant data should not be gathered to prevent potential misuse. Data protection and governance frameworks should be set up and adhered to by developers and deployers of AI systems. These frameworks should also be periodically reviewed and updated in accordance with applicable privacy and data protection laws. For example, data protection impact assessments (DPIA) help organisations determine how data processing systems, procedures, or technologies affect individuals’ privacy and eliminate risks that might violate compliance7. However, it is important to note that DPIAs are much narrower in scope than an overall impact assessment for use of AI systems and are not sufficient as an AI risk assessment. Other components will need to be considered for a full assessment of risks associated with AI systems. Developers and deployers of AI systems should also incorporate a privacy by design principle when developing and deploying AI systems. Privacy by design is an approach that embeds privacy in every stage of the system development lifecycle. Data privacy is essential in gaining the public’s trust in technological advances. Another consideration is investing in privacy enhancing technologies to preserve privacy while allowing personal data to be used for innovation. Privacy enhancing technologies include, but are not limited to, differential privacy, where small changes are made to raw data to securely de identify inputs without having a significant impact on the results of the AI system, and zero knowledge proofs (ZKP), where ZKP hide the underlying data and answer simple questions about whether something is true or false without revealing additional information

Published by ASEAN in ASEAN Guide on AI Governance and Ethics, 2024

2

In domestic regulation, we recommend mandatory registration for the creation, sale or use of models above a certain capability threshold, including open source copies and derivatives, to enable governments to acquire critical and currently missing visibility into emerging risks. Governments should monitor large scale data centers and track AI incidents, and should require that AI developers of frontier models be subject to independent third party audits evaluating their information security and model safety. AI developers should also be required to share comprehensive risk assessments, policies around risk management, and predictions about their systems’ behaviour in third party evaluations and post deployment with relevant authorities.

Published by IDAIS (International Dialogues on AI Safety) in IDAIS-Oxford, Oct 31, 2023

3. Auditability

Enable interested third parties to probe, understand, and review the behaviour of the algorithm through disclosure of information that enables monitoring, checking or criticism.

Published by Personal Data Protection Commission (PDPC), Singapore in A compilation of existing AI ethical principles (Annex A), Jan 21, 2020

· Prepare Input Data:

1 The exercise of data procurement, management, and organization should uphold the legal frameworks and standards of data privacy. Data privacy and security protect information from a wide range of threats. 2 The confidentiality of data ensures that information is accessible only to those who are authorized to access the information and that there are specific controls that manage the delegation of authority. 3 Designers and engineers of the AI system must exhibit the appropriate levels of integrity to safeguard the accuracy and completeness of information and processing methods to ensure that the privacy and security legal framework and standards are followed. They should also ensure that the availability and storage of data are protected through suitable security database systems. 4 All processed data should be classified to ensure that it receives the appropriate level of protection in accordance with its sensitivity or security classification and that AI system developers and owners are aware of the classification or sensitivity of the information they are handling and the associated requirements to keep it secure. All data shall be classified in terms of business requirements, criticality, and sensitivity in order to prevent unauthorized disclosure or modification. Data classification should be conducted in a contextual manner that does not result in the inference of personal information. Furthermore, de identification mechanisms should be employed based on data classification as well as requirements relating to data protection laws. 5 Data backups and archiving actions should be taken in this stage to align with business continuity, disaster recovery and risk mitigation policies.

Published by SDAIA in AI Ethics Principles, Sept 14, 2022

3 Ensure transparency, explainability and intelligibility

AI should be intelligible or understandable to developers, users and regulators. Two broad approaches to ensuring intelligibility are improving the transparency and explainability of AI technology. Transparency requires that sufficient information (described below) be published or documented before the design and deployment of an AI technology. Such information should facilitate meaningful public consultation and debate on how the AI technology is designed and how it should be used. Such information should continue to be published and documented regularly and in a timely manner after an AI technology is approved for use. Transparency will improve system quality and protect patient and public health safety. For instance, system evaluators require transparency in order to identify errors, and government regulators rely on transparency to conduct proper, effective oversight. It must be possible to audit an AI technology, including if something goes wrong. Transparency should include accurate information about the assumptions and limitations of the technology, operating protocols, the properties of the data (including methods of data collection, processing and labelling) and development of the algorithmic model. AI technologies should be explainable to the extent possible and according to the capacity of those to whom the explanation is directed. Data protection laws already create specific obligations of explainability for automated decision making. Those who might request or require an explanation should be well informed, and the educational information must be tailored to each population, including, for example, marginalized populations. Many AI technologies are complex, and the complexity might frustrate both the explainer and the person receiving the explanation. There is a possible trade off between full explainability of an algorithm (at the cost of accuracy) and improved accuracy (at the cost of explainability). All algorithms should be tested rigorously in the settings in which the technology will be used in order to ensure that it meets standards of safety and efficacy. The examination and validation should include the assumptions, operational protocols, data properties and output decisions of the AI technology. Tests and evaluations should be regular, transparent and of sufficient breadth to cover differences in the performance of the algorithm according to race, ethnicity, gender, age and other relevant human characteristics. There should be robust, independent oversight of such tests and evaluation to ensure that they are conducted safely and effectively. Health care institutions, health systems and public health agencies should regularly publish information about how decisions have been made for adoption of an AI technology and how the technology will be evaluated periodically, its uses, its known limitations and the role of decision making, which can facilitate external auditing and oversight.

Published by World Health Organization (WHO) in Key ethical principles for use of artificial intelligence for health, Jun 28, 2021