Australia does not have a single AI fairness statute. Unfair or biased AI outcomes can nevertheless engage Commonwealth, state or territory anti-discrimination laws, employment law, consumer protection, privacy, credit, education, health, administrative law and other sector-specific duties. The applicable protected attributes, tests, exemptions, remedies and responsible parties depend on the relevant statute and context.

The current Guidance for AI adoption addresses fairness through impact analysis, stakeholder engagement, risk management, testing, monitoring and human control. The OAIC identifies bias and discrimination risks where data are incomplete, inaccurate, unrepresentative or encode historical disadvantage. These are guidance propositions unless linked to a specific legal obligation.

As a governance matter, organisations deploying higher-impact AI should test for discriminatory or materially inaccurate outcomes before and after deployment; assess performance across relevant cohorts; document limitations; monitor complaints and drift; maintain escalation and contestability pathways; and ensure that human reviewers have the authority and information needed to correct inappropriate outcomes.

At its core, the EU AI Act is driven by the imperative to safeguard the fundamental rights of EU citizens. The rapid advancement of AI technologies has introduced significant benefits but also potential risks, such as biases in decision-making systems and privacy infringements. The AI Act aims to mitigate these risks by establishing clear rules that ensure AI systems respect the rights enshrined in the EU Charter of Fundamental Rights. This focus on human-centric AI seeks to enhance trust and acceptance among the public, thereby promoting wider adoption of AI technologies in a responsible manner.

Within the EU AI Act, non-discrimination and fairness is incorporated withing the following:

• Recital 27 includes seven principles for trustworthy AI including ensuring that AI systems are developed and used in a way that includes diverse actors and promotes equal access, gender equality and cultural diversity, while avoiding discriminatory impacts and unfair biases that are prohibited by Union or national law.
• Article 10 sets out data and data governance requirements for high-risk AI systems and includes a requirement to examine and assess possible bias in training, validation and testing data sets.
• Deployers are required to ensure that any input data is relevant and sufficiently representative in view of the intended purpose of the high-risk AI system (Article 26(4)).

The Framework addresses the issue of bias (most notably in paragraphs 27-37 relating to ‘Non-bias and non-discrimination') and highlighted that AI has the potential to create and reinforce biases and that bias and discrimination by AI can cause manifest harm to individuals and to society. The European Parliament stated that regulation should encourage the development and sharing of strategies to counter these risks, including debiasing datasets in research and development and by the development of rules on data processing. The European Parliament also considered this approach to have the potential to turn software, algorithms and data into an asset in fighting bias and discrimination in certain situations, and a force for equal rights and positive social change.

At its core, the EU AI Act is driven by the imperative to safeguard the fundamental rights of EU citizens. The rapid advancement of AI technologies has introduced significant benefits but also potential risks, such as biases in decision-making systems and privacy infringements. The AI Act aims to mitigate these risks by establishing clear rules that ensure AI systems respect the rights enshrined in the EU Charter of Fundamental Rights. This focus on human-centric AI seeks to enhance trust and acceptance among the public, thereby promoting wider adoption of AI technologies in a responsible manner.

Within the EU AI Act, non-discrimination and fairness is incorporated withing the following:

• Recital 27 includes seven principles for trustworthy AI including ensuring that AI systems are developed and used in a way that includes diverse actors and promotes equal access, gender equality and cultural diversity, while avoiding discriminatory impacts and unfair biases that are prohibited by Union or national law.
• Article 10 sets out data and data governance requirements for high-risk AI systems and includes a requirement to examine and assess possible bias in training, validation and testing data sets.
• Deployers are required to ensure that any input data is relevant and sufficiently representative in view of the intended purpose of the high-risk AI system (Article 26(4)).

The Framework addresses the issue of bias (most notably in paragraphs 27-37 relating to ‘Non-bias and non-discrimination') and highlighted that AI has the potential to create and reinforce biases and that bias and discrimination by AI can cause manifest harm to individuals and to society. The European Parliament stated that regulation should encourage the development and sharing of strategies to counter these risks, including debiasing datasets in research and development and by the development of rules on data processing. The European Parliament also considered this approach to have the potential to turn software, algorithms and data into an asset in fighting bias and discrimination in certain situations, and a force for equal rights and positive social change.

The Brazilian AI Strategy discusses the importance of establishing mechanisms that allow the prevention and elimination of biases, which can result both from the algorithms used as well as from the databases used for their training (Section 2 – ‘Governance of AI’, page 24, Brazilian AI Strategy).

At its core, the EU AI Act is driven by the imperative to safeguard the fundamental rights of EU citizens. The rapid advancement of AI technologies has introduced significant benefits but also potential risks, such as biases in decision-making systems and privacy infringements. The AI Act aims to mitigate these risks by establishing clear rules that ensure AI systems respect the rights enshrined in the EU Charter of Fundamental Rights. This focus on human-centric AI seeks to enhance trust and acceptance among the public, thereby promoting wider adoption of AI technologies in a responsible manner.

Within the EU AI Act, non-discrimination and fairness is incorporated withing the following:

• Recital 27 includes seven principles for trustworthy AI including ensuring that AI systems are developed and used in a way that includes diverse actors and promotes equal access, gender equality and cultural diversity, while avoiding discriminatory impacts and unfair biases that are prohibited by Union or national law.
• Article 10 sets out data and data governance requirements for high-risk AI systems and includes a requirement to examine and assess possible bias in training, validation and testing data sets.
• Deployers are required to ensure that any input data is relevant and sufficiently representative in view of the intended purpose of the high-risk AI system (Article 26(4)).

The Framework addresses the issue of bias (most notably in paragraphs 27-37 relating to ‘Non-bias and non-discrimination') and highlighted that AI has the potential to create and reinforce biases and that bias and discrimination by AI can cause manifest harm to individuals and to society. The European Parliament stated that regulation should encourage the development and sharing of strategies to counter these risks, including debiasing datasets in research and development and by the development of rules on data processing. The European Parliament also considered this approach to have the potential to turn software, algorithms and data into an asset in fighting bias and discrimination in certain situations, and a force for equal rights and positive social change.

The Voluntary Code specifies under its Fairness and Equity principle that signatories should (with varying levels of obligation, as indicated, depending on whether a signatory is either a developer or a manager of a generative AI system and if the system is available for public use or not):

• assess and curate datasets used for training to manage data quality and potential biases; and
• implement diverse testing methods and measures to assess and mitigate risk of biased output prior to release.

Article 4 establishes the main principles applicable to AI systems, and Article 4 e) states the following:

Diversity, non-discrimination and equity

Diversity, non-discrimination and equity: AI systems will be developed and used throughout their lifecycle, promoting equal access, gender equality and cultural diversity, whilst avoiding discriminatory effects and selection or information biases that could generate a discriminatory effect.

The GenAI Measures require that measures be taken to prevent discrimination on the basis of race, ethnicity, beliefs, nationality, region, gender, age, occupation, etc. in the process of algorithm design, training data selection, model generation and optimisation and provision of services. Further, the lawful rights and interests of others (including rights to likeness, reputation, honour, personal privacy and personal information) must also be respected.

Under the Recommendation Algorithms Provisions, service providers have a responsibility to safeguard specific protected groups, including minors and the elderly, by providing appropriate services that are in line with such groups' characteristics. Where a recommendation algorithm-based service is deployed in an employee work dispatching use case, they must also ensure workers' rights to compensation, rest and leave. Consumers' right to fair trading must also be protected where a service is deployed to provide goods or services to consumers.

The Trial Measures for Ethical Review and Service of Artificial Intelligence Technology (issued April 2026) further address fairness and bias by requiring AI ethics reviews to focus on fairness and justice, and by addressing specific issues such as training data selection criteria, algorithm rationality and measures to prevent bias and discrimination.

At its core, the EU AI Act is driven by the imperative to safeguard the fundamental rights of EU citizens. The rapid advancement of AI technologies has introduced significant benefits but also potential risks, such as biases in decision-making systems and privacy infringements. The AI Act aims to mitigate these risks by establishing clear rules that ensure AI systems respect the rights enshrined in the EU Charter of Fundamental Rights. This focus on human-centric AI seeks to enhance trust and acceptance among the public, thereby promoting wider adoption of AI technologies in a responsible manner.

Within the EU AI Act, non-discrimination and fairness is incorporated withing the following:

• Recital 27 includes seven principles for trustworthy AI including ensuring that AI systems are developed and used in a way that includes diverse actors and promotes equal access, gender equality and cultural diversity, while avoiding discriminatory impacts and unfair biases that are prohibited by Union or national law.
• Article 10 sets out data and data governance requirements for high-risk AI systems and includes a requirement to examine and assess possible bias in training, validation and testing data sets.
• Deployers are required to ensure that any input data is relevant and sufficiently representative in view of the intended purpose of the high-risk AI system (Article 26(4)).

The Framework addresses the issue of bias (most notably in paragraphs 27-37 relating to ‘Non-bias and non-discrimination') and highlighted that AI has the potential to create and reinforce biases and that bias and discrimination by AI can cause manifest harm to individuals and to society. The European Parliament stated that regulation should encourage the development and sharing of strategies to counter these risks, including debiasing datasets in research and development and by the development of rules on data processing. The European Parliament also considered this approach to have the potential to turn software, algorithms and data into an asset in fighting bias and discrimination in certain situations, and a force for equal rights and positive social change.

At its core, the EU AI Act is driven by the imperative to safeguard the fundamental rights of EU citizens. The rapid advancement of AI technologies has introduced significant benefits but also potential risks, such as biases in decision-making systems and privacy infringements. The AI Act aims to mitigate these risks by establishing clear rules that ensure AI systems respect the rights enshrined in the EU Charter of Fundamental Rights. This focus on human-centric AI seeks to enhance trust and acceptance among the public, thereby promoting wider adoption of AI technologies in a responsible manner.

Within the EU AI Act, non-discrimination and fairness is incorporated withing the following:

• Recital 27 includes seven principles for trustworthy AI including ensuring that AI systems are developed and used in a way that includes diverse actors and promotes equal access, gender equality and cultural diversity, while avoiding discriminatory impacts and unfair biases that are prohibited by Union or national law.
• Article 10 sets out data and data governance requirements for high-risk AI systems and includes a requirement to examine and assess possible bias in training, validation and testing data sets.
• Deployers are required to ensure that any input data is relevant and sufficiently representative in view of the intended purpose of the high-risk AI system (Article 26(4)).

The Framework addresses the issue of bias (most notably in paragraphs 27-37 relating to ‘Non-bias and non-discrimination') and highlighted that AI has the potential to create and reinforce biases and that bias and discrimination by AI can cause manifest harm to individuals and to society. The European Parliament stated that regulation should encourage the development and sharing of strategies to counter these risks, including debiasing datasets in research and development and by the development of rules on data processing. The European Parliament also considered this approach to have the potential to turn software, algorithms and data into an asset in fighting bias and discrimination in certain situations, and a force for equal rights and positive social change.

At its core, the EU AI Act is driven by the imperative to safeguard the fundamental rights of EU citizens. The rapid advancement of AI technologies has introduced significant benefits but also potential risks, such as biases in decision-making systems and privacy infringements. The AI Act aims to mitigate these risks by establishing clear rules that ensure AI systems respect the rights enshrined in the EU Charter of Fundamental Rights. This focus on human-centric AI seeks to enhance trust and acceptance among the public, thereby promoting wider adoption of AI technologies in a responsible manner.

Within the EU AI Act, non-discrimination and fairness is incorporated withing the following:

• Recital 27 includes seven principles for trustworthy AI including ensuring that AI systems are developed and used in a way that includes diverse actors and promotes equal access, gender equality and cultural diversity, while avoiding discriminatory impacts and unfair biases that are prohibited by Union or national law.
• Article 10 sets out data and data governance requirements for high-risk AI systems and includes a requirement to examine and assess possible bias in training, validation and testing data sets.
• Deployers are required to ensure that any input data is relevant and sufficiently representative in view of the intended purpose of the high-risk AI system (Article 26(4)).

The Framework addresses the issue of bias (most notably in paragraphs 27-37 relating to ‘Non-bias and non-discrimination') and highlighted that AI has the potential to create and reinforce biases and that bias and discrimination by AI can cause manifest harm to individuals and to society. The European Parliament stated that regulation should encourage the development and sharing of strategies to counter these risks, including debiasing datasets in research and development and by the development of rules on data processing. The European Parliament also considered this approach to have the potential to turn software, algorithms and data into an asset in fighting bias and discrimination in certain situations, and a force for equal rights and positive social change.

At its core, the EU AI Act is driven by the imperative to safeguard the fundamental rights of EU citizens. The rapid advancement of AI technologies has introduced significant benefits but also potential risks, such as biases in decision-making systems and privacy infringements. The AI Act aims to mitigate these risks by establishing clear rules that ensure AI systems respect the rights enshrined in the EU Charter of Fundamental Rights. This focus on human-centric AI seeks to enhance trust and acceptance among the public, thereby promoting wider adoption of AI technologies in a responsible manner.

Within the EU AI Act, non-discrimination and fairness is incorporated withing the following:

• Recital 27 includes seven principles for trustworthy AI including ensuring that AI systems are developed and used in a way that includes diverse actors and promotes equal access, gender equality and cultural diversity, while avoiding discriminatory impacts and unfair biases that are prohibited by Union or national law.
• Article 10 sets out data and data governance requirements for high-risk AI systems and includes a requirement to examine and assess possible bias in training, validation and testing data sets.
• Deployers are required to ensure that any input data is relevant and sufficiently representative in view of the intended purpose of the high-risk AI system (Article 26(4)).

The Framework addresses the issue of bias (most notably in paragraphs 27-37 relating to ‘Non-bias and non-discrimination') and highlighted that AI has the potential to create and reinforce biases and that bias and discrimination by AI can cause manifest harm to individuals and to society. The European Parliament stated that regulation should encourage the development and sharing of strategies to counter these risks, including debiasing datasets in research and development and by the development of rules on data processing. The European Parliament also considered this approach to have the potential to turn software, algorithms and data into an asset in fighting bias and discrimination in certain situations, and a force for equal rights and positive social change.

At its core, the EU AI Act is driven by the imperative to safeguard the fundamental rights of EU citizens. The rapid advancement of AI technologies has introduced significant benefits but also potential risks, such as biases in decision-making systems and privacy infringements. The AI Act aims to mitigate these risks by establishing clear rules that ensure AI systems respect the rights enshrined in the EU Charter of Fundamental Rights. This focus on human-centric AI seeks to enhance trust and acceptance among the public, thereby promoting wider adoption of AI technologies in a responsible manner.

Within the EU AI Act, non-discrimination and fairness is incorporated withing the following:

• Recital 27 includes seven principles for trustworthy AI including ensuring that AI systems are developed and used in a way that includes diverse actors and promotes equal access, gender equality and cultural diversity, while avoiding discriminatory impacts and unfair biases that are prohibited by Union or national law.
• Article 10 sets out data and data governance requirements for high-risk AI systems and includes a requirement to examine and assess possible bias in training, validation and testing data sets.
• Deployers are required to ensure that any input data is relevant and sufficiently representative in view of the intended purpose of the high-risk AI system (Article 26(4)).

The Framework addresses the issue of bias (most notably in paragraphs 27-37 relating to ‘Non-bias and non-discrimination') and highlighted that AI has the potential to create and reinforce biases and that bias and discrimination by AI can cause manifest harm to individuals and to society. The European Parliament stated that regulation should encourage the development and sharing of strategies to counter these risks, including debiasing datasets in research and development and by the development of rules on data processing. The European Parliament also considered this approach to have the potential to turn software, algorithms and data into an asset in fighting bias and discrimination in certain situations, and a force for equal rights and positive social change.

At its core, the EU AI Act is driven by the imperative to safeguard the fundamental rights of EU citizens. The rapid advancement of AI technologies has introduced significant benefits but also potential risks, such as biases in decision-making systems and privacy infringements. The AI Act aims to mitigate these risks by establishing clear rules that ensure AI systems respect the rights enshrined in the EU Charter of Fundamental Rights. This focus on human-centric AI seeks to enhance trust and acceptance among the public, thereby promoting wider adoption of AI technologies in a responsible manner.

Within the EU AI Act, non-discrimination and fairness is incorporated withing the following:

• Recital 27 includes seven principles for trustworthy AI including ensuring that AI systems are developed and used in a way that includes diverse actors and promotes equal access, gender equality and cultural diversity, while avoiding discriminatory impacts and unfair biases that are prohibited by Union or national law.
• Article 10 sets out data and data governance requirements for high-risk AI systems and includes a requirement to examine and assess possible bias in training, validation and testing data sets.
• Deployers are required to ensure that any input data is relevant and sufficiently representative in view of the intended purpose of the high-risk AI system (Article 26(4)).

The Framework addresses the issue of bias (most notably in paragraphs 27-37 relating to ‘Non-bias and non-discrimination') and highlighted that AI has the potential to create and reinforce biases and that bias and discrimination by AI can cause manifest harm to individuals and to society. The European Parliament stated that regulation should encourage the development and sharing of strategies to counter these risks, including debiasing datasets in research and development and by the development of rules on data processing. The European Parliament also considered this approach to have the potential to turn software, algorithms and data into an asset in fighting bias and discrimination in certain situations, and a force for equal rights and positive social change.

At its core, the EU AI Act is driven by the imperative to safeguard the fundamental rights of EU citizens. The rapid advancement of AI technologies has introduced significant benefits but also potential risks, such as biases in decision-making systems and privacy infringements. The AI Act aims to mitigate these risks by establishing clear rules that ensure AI systems respect the rights enshrined in the EU Charter of Fundamental Rights. This focus on human-centric AI seeks to enhance trust and acceptance among the public, thereby promoting wider adoption of AI technologies in a responsible manner.

Within the EU AI Act, non-discrimination and fairness is incorporated withing the following:

• Recital 27 includes seven principles for trustworthy AI including ensuring that AI systems are developed and used in a way that includes diverse actors and promotes equal access, gender equality and cultural diversity, while avoiding discriminatory impacts and unfair biases that are prohibited by Union or national law.
• Article 10 sets out data and data governance requirements for high-risk AI systems and includes a requirement to examine and assess possible bias in training, validation and testing data sets.
• Deployers are required to ensure that any input data is relevant and sufficiently representative in view of the intended purpose of the high-risk AI system (Article 26(4)).

The Framework addresses the issue of bias (most notably in paragraphs 27-37 relating to ‘Non-bias and non-discrimination') and highlighted that AI has the potential to create and reinforce biases and that bias and discrimination by AI can cause manifest harm to individuals and to society. The European Parliament stated that regulation should encourage the development and sharing of strategies to counter these risks, including debiasing datasets in research and development and by the development of rules on data processing. The European Parliament also considered this approach to have the potential to turn software, algorithms and data into an asset in fighting bias and discrimination in certain situations, and a force for equal rights and positive social change.

At its core, the EU AI Act is driven by the imperative to safeguard the fundamental rights of EU citizens. The rapid advancement of AI technologies has introduced significant benefits but also potential risks, such as biases in decision-making systems and privacy infringements. The AI Act aims to mitigate these risks by establishing clear rules that ensure AI systems respect the rights enshrined in the EU Charter of Fundamental Rights. This focus on human-centric AI seeks to enhance trust and acceptance among the public, thereby promoting wider adoption of AI technologies in a responsible manner.

Within the EU AI Act, non-discrimination and fairness is incorporated withing the following:

• Recital 27 includes seven principles for trustworthy AI including ensuring that AI systems are developed and used in a way that includes diverse actors and promotes equal access, gender equality and cultural diversity, while avoiding discriminatory impacts and unfair biases that are prohibited by Union or national law.
• Article 10 sets out data and data governance requirements for high-risk AI systems and includes a requirement to examine and assess possible bias in training, validation and testing data sets.
• Deployers are required to ensure that any input data is relevant and sufficiently representative in view of the intended purpose of the high-risk AI system (Article 26(4)).

The Framework addresses the issue of bias (most notably in paragraphs 27-37 relating to ‘Non-bias and non-discrimination') and highlighted that AI has the potential to create and reinforce biases and that bias and discrimination by AI can cause manifest harm to individuals and to society. The European Parliament stated that regulation should encourage the development and sharing of strategies to counter these risks, including debiasing datasets in research and development and by the development of rules on data processing. The European Parliament also considered this approach to have the potential to turn software, algorithms and data into an asset in fighting bias and discrimination in certain situations, and a force for equal rights and positive social change.

At its core, the EU AI Act is driven by the imperative to safeguard the fundamental rights of EU citizens. The rapid advancement of AI technologies has introduced significant benefits but also potential risks, such as biases in decision-making systems and privacy infringements. The AI Act aims to mitigate these risks by establishing clear rules that ensure AI systems respect the rights enshrined in the EU Charter of Fundamental Rights. This focus on human-centric AI seeks to enhance trust and acceptance among the public, thereby promoting wider adoption of AI technologies in a responsible manner.

Within the EU AI Act, non-discrimination and fairness is incorporated withing the following:

• Recital 27 includes seven principles for trustworthy AI including ensuring that AI systems are developed and used in a way that includes diverse actors and promotes equal access, gender equality and cultural diversity, while avoiding discriminatory impacts and unfair biases that are prohibited by Union or national law.
• Article 10 sets out data and data governance requirements for high-risk AI systems and includes a requirement to examine and assess possible bias in training, validation and testing data sets.
• Deployers are required to ensure that any input data is relevant and sufficiently representative in view of the intended purpose of the high-risk AI system (Article 26(4)).

The Framework addresses the issue of bias (most notably in paragraphs 27-37 relating to ‘Non-bias and non-discrimination') and highlighted that AI has the potential to create and reinforce biases and that bias and discrimination by AI can cause manifest harm to individuals and to society. The European Parliament stated that regulation should encourage the development and sharing of strategies to counter these risks, including debiasing datasets in research and development and by the development of rules on data processing. The European Parliament also considered this approach to have the potential to turn software, algorithms and data into an asset in fighting bias and discrimination in certain situations, and a force for equal rights and positive social change.

At its core, the EU AI Act is driven by the imperative to safeguard the fundamental rights of EU citizens. The rapid advancement of AI technologies has introduced significant benefits but also potential risks, such as biases in decision-making systems and privacy infringements. The AI Act aims to mitigate these risks by establishing clear rules that ensure AI systems respect the rights enshrined in the EU Charter of Fundamental Rights. This focus on human-centric AI seeks to enhance trust and acceptance among the public, thereby promoting wider adoption of AI technologies in a responsible manner.

Within the EU AI Act, non-discrimination and fairness is incorporated withing the following:

• Recital 27 includes seven principles for trustworthy AI including ensuring that AI systems are developed and used in a way that includes diverse actors and promotes equal access, gender equality and cultural diversity, while avoiding discriminatory impacts and unfair biases that are prohibited by Union or national law.
• Article 10 sets out data and data governance requirements for high-risk AI systems and includes a requirement to examine and assess possible bias in training, validation and testing data sets.
• Deployers are required to ensure that any input data is relevant and sufficiently representative in view of the intended purpose of the high-risk AI system (Article 26(4)).

The Framework addresses the issue of bias (most notably in paragraphs 27-37 relating to ‘Non-bias and non-discrimination') and highlighted that AI has the potential to create and reinforce biases and that bias and discrimination by AI can cause manifest harm to individuals and to society. The European Parliament stated that regulation should encourage the development and sharing of strategies to counter these risks, including debiasing datasets in research and development and by the development of rules on data processing. The European Parliament also considered this approach to have the potential to turn software, algorithms and data into an asset in fighting bias and discrimination in certain situations, and a force for equal rights and positive social change.

Laws specifically addressing AI have not yet been introduced in Hong Kong. However, existing anti-discrimination laws, namely the Sex Discrimination Ordinance (Cap. 480), Disability Discrimination Ordinance (Cap. 487), Family Status Discrimination Ordinance (Cap. 527), and the Race Discrimination Ordinance (Cap. 602) remain fully applicable against bias or discrimination in AI contexts.

The GenAI Guideline recommends risk assessments and strict controls to be implemented at each stage (from initial data collection, model training and content generation) to mitigate model biases. It recommends continuous monitoring and audits to identify and address bias.

The fairness principle within the Ethical AI Framework expects recommendations/results from the AI application to treat individuals within similar groups in a fair manner, without favouritism or discrimination and without causing or resulting in harm. It further provides that this entails maintaining respect for the individuals behind the data and refraining from using datasets that contain discriminatory biases. It recommends various measures for mitigating these risks.

The fairness ethical principle set out in the Guidance specifies that individuals are entitled to be treated in a reasonably equal manner, without unjust bias or unlawful discrimination, and that differential treatments between different individuals or different groups of people should be justifiable with sound reasons. The Model Framework expands on this, including recommending certain measures to mitigate these risks, such as validation and testing.

At its core, the EU AI Act is driven by the imperative to safeguard the fundamental rights of EU citizens. The rapid advancement of AI technologies has introduced significant benefits but also potential risks, such as biases in decision-making systems and privacy infringements. The AI Act aims to mitigate these risks by establishing clear rules that ensure AI systems respect the rights enshrined in the EU Charter of Fundamental Rights. This focus on human-centric AI seeks to enhance trust and acceptance among the public, thereby promoting wider adoption of AI technologies in a responsible manner.

Within the EU AI Act, non-discrimination and fairness is incorporated withing the following:

• Recital 27 includes seven principles for trustworthy AI including ensuring that AI systems are developed and used in a way that includes diverse actors and promotes equal access, gender equality and cultural diversity, while avoiding discriminatory impacts and unfair biases that are prohibited by Union or national law.
• Article 10 sets out data and data governance requirements for high-risk AI systems and includes a requirement to examine and assess possible bias in training, validation and testing data sets.
• Deployers are required to ensure that any input data is relevant and sufficiently representative in view of the intended purpose of the high-risk AI system (Article 26(4)).

The Framework addresses the issue of bias (most notably in paragraphs 27-37 relating to ‘Non-bias and non-discrimination') and highlighted that AI has the potential to create and reinforce biases and that bias and discrimination by AI can cause manifest harm to individuals and to society. The European Parliament stated that regulation should encourage the development and sharing of strategies to counter these risks, including debiasing datasets in research and development and by the development of rules on data processing. The European Parliament also considered this approach to have the potential to turn software, algorithms and data into an asset in fighting bias and discrimination in certain situations, and a force for equal rights and positive social change.

India currently lacks legislation specific to AI. Consequently, there is no codified framework for addressing fairness or anti‑bias principles within a single regulation. Although the sutras of Trust and Fairness and Equity forming part of the India AI Governance Guidelines and certain sector-specific frameworks emphasise the requirement of fairness and mitigating unlawful bias.

At its core, the EU AI Act is driven by the imperative to safeguard the fundamental rights of EU citizens. The rapid advancement of AI technologies has introduced significant benefits but also potential risks, such as biases in decision-making systems and privacy infringements. The AI Act aims to mitigate these risks by establishing clear rules that ensure AI systems respect the rights enshrined in the EU Charter of Fundamental Rights. This focus on human-centric AI seeks to enhance trust and acceptance among the public, thereby promoting wider adoption of AI technologies in a responsible manner.

Within the EU AI Act, non-discrimination and fairness is incorporated withing the following:

• Recital 27 includes seven principles for trustworthy AI including ensuring that AI systems are developed and used in a way that includes diverse actors and promotes equal access, gender equality and cultural diversity, while avoiding discriminatory impacts and unfair biases that are prohibited by Union or national law.
• Article 10 sets out data and data governance requirements for high-risk AI systems and includes a requirement to examine and assess possible bias in training, validation and testing data sets.
• Deployers are required to ensure that any input data is relevant and sufficiently representative in view of the intended purpose of the high-risk AI system (Article 26(4)).

The Framework addresses the issue of bias (most notably in paragraphs 27-37 relating to ‘Non-bias and non-discrimination') and highlighted that AI has the potential to create and reinforce biases and that bias and discrimination by AI can cause manifest harm to individuals and to society. The European Parliament stated that regulation should encourage the development and sharing of strategies to counter these risks, including debiasing datasets in research and development and by the development of rules on data processing. The European Parliament also considered this approach to have the potential to turn software, algorithms and data into an asset in fighting bias and discrimination in certain situations, and a force for equal rights and positive social change.

There is no AI‑specific anti‑bias statute. Existing laws continue to apply to AI‑enabled conduct, including statutory prohibitions on discrimination.

In that spirit, the 2023 AI Policy Paper calls for fairness and non‑discrimination in AI systems, and the Financial Sector Report underscores the risk that such systems may perpetuate discrimination through biased training data or proxy variables. To mitigate these risks, the Financial Sector Report recommends, among other measures, continuous oversight throughout the AI lifecycle and outcome-based testing designed to detect and address discriminatory effects.

At its core, the EU AI Act is driven by the imperative to safeguard the fundamental rights of EU citizens. The rapid advancement of AI technologies has introduced significant benefits but also potential risks, such as biases in decision-making systems and privacy infringements. The AI Act aims to mitigate these risks by establishing clear rules that ensure AI systems respect the rights enshrined in the EU Charter of Fundamental Rights. This focus on human-centric AI seeks to enhance trust and acceptance among the public, thereby promoting wider adoption of AI technologies in a responsible manner.

Within the EU AI Act, non-discrimination and fairness is incorporated withing the following:

• Recital 27 includes seven principles for trustworthy AI including ensuring that AI systems are developed and used in a way that includes diverse actors and promotes equal access, gender equality and cultural diversity, while avoiding discriminatory impacts and unfair biases that are prohibited by Union or national law.
• Article 10 sets out data and data governance requirements for high-risk AI systems and includes a requirement to examine and assess possible bias in training, validation and testing data sets.
• Deployers are required to ensure that any input data is relevant and sufficiently representative in view of the intended purpose of the high-risk AI system (Article 26(4)).

The Framework addresses the issue of bias (most notably in paragraphs 27-37 relating to ‘Non-bias and non-discrimination') and highlighted that AI has the potential to create and reinforce biases and that bias and discrimination by AI can cause manifest harm to individuals and to society. The European Parliament stated that regulation should encourage the development and sharing of strategies to counter these risks, including debiasing datasets in research and development and by the development of rules on data processing. The European Parliament also considered this approach to have the potential to turn software, algorithms and data into an asset in fighting bias and discrimination in certain situations, and a force for equal rights and positive social change.

The Social Principles state that the use of AI should not create inequality or social disadvantage. Relevant policy makers and businesses must have a thorough understanding of AI, along with the knowledge and ethical awareness to use AI appropriately. Furthermore, an educational environment that promotes learning and literacy must be made accessible to everyone.

At its core, the EU AI Act is driven by the imperative to safeguard the fundamental rights of EU citizens. The rapid advancement of AI technologies has introduced significant benefits but also potential risks, such as biases in decision-making systems and privacy infringements. The AI Act aims to mitigate these risks by establishing clear rules that ensure AI systems respect the rights enshrined in the EU Charter of Fundamental Rights. This focus on human-centric AI seeks to enhance trust and acceptance among the public, thereby promoting wider adoption of AI technologies in a responsible manner.

Within the EU AI Act, non-discrimination and fairness is incorporated withing the following:

• Recital 27 includes seven principles for trustworthy AI including ensuring that AI systems are developed and used in a way that includes diverse actors and promotes equal access, gender equality and cultural diversity, while avoiding discriminatory impacts and unfair biases that are prohibited by Union or national law.
• Article 10 sets out data and data governance requirements for high-risk AI systems and includes a requirement to examine and assess possible bias in training, validation and testing data sets.
• Deployers are required to ensure that any input data is relevant and sufficiently representative in view of the intended purpose of the high-risk AI system (Article 26(4)).

The Framework addresses the issue of bias (most notably in paragraphs 27-37 relating to ‘Non-bias and non-discrimination') and highlighted that AI has the potential to create and reinforce biases and that bias and discrimination by AI can cause manifest harm to individuals and to society. The European Parliament stated that regulation should encourage the development and sharing of strategies to counter these risks, including debiasing datasets in research and development and by the development of rules on data processing. The European Parliament also considered this approach to have the potential to turn software, algorithms and data into an asset in fighting bias and discrimination in certain situations, and a force for equal rights and positive social change.

At its core, the EU AI Act is driven by the imperative to safeguard the fundamental rights of EU citizens. The rapid advancement of AI technologies has introduced significant benefits but also potential risks, such as biases in decision-making systems and privacy infringements. The AI Act aims to mitigate these risks by establishing clear rules that ensure AI systems respect the rights enshrined in the EU Charter of Fundamental Rights. This focus on human-centric AI seeks to enhance trust and acceptance among the public, thereby promoting wider adoption of AI technologies in a responsible manner.

Within the EU AI Act, non-discrimination and fairness is incorporated withing the following:

• Recital 27 includes seven principles for trustworthy AI including ensuring that AI systems are developed and used in a way that includes diverse actors and promotes equal access, gender equality and cultural diversity, while avoiding discriminatory impacts and unfair biases that are prohibited by Union or national law.
• Article 10 sets out data and data governance requirements for high-risk AI systems and includes a requirement to examine and assess possible bias in training, validation and testing data sets.
• Deployers are required to ensure that any input data is relevant and sufficiently representative in view of the intended purpose of the high-risk AI system (Article 26(4)).

The Framework addresses the issue of bias (most notably in paragraphs 27-37 relating to ‘Non-bias and non-discrimination') and highlighted that AI has the potential to create and reinforce biases and that bias and discrimination by AI can cause manifest harm to individuals and to society. The European Parliament stated that regulation should encourage the development and sharing of strategies to counter these risks, including debiasing datasets in research and development and by the development of rules on data processing. The European Parliament also considered this approach to have the potential to turn software, algorithms and data into an asset in fighting bias and discrimination in certain situations, and a force for equal rights and positive social change.

At its core, the EU AI Act is driven by the imperative to safeguard the fundamental rights of EU citizens. The rapid advancement of AI technologies has introduced significant benefits but also potential risks, such as biases in decision-making systems and privacy infringements. The AI Act aims to mitigate these risks by establishing clear rules that ensure AI systems respect the rights enshrined in the EU Charter of Fundamental Rights. This focus on human-centric AI seeks to enhance trust and acceptance among the public, thereby promoting wider adoption of AI technologies in a responsible manner.

Within the EU AI Act, non-discrimination and fairness is incorporated withing the following:

• Recital 27 includes seven principles for trustworthy AI including ensuring that AI systems are developed and used in a way that includes diverse actors and promotes equal access, gender equality and cultural diversity, while avoiding discriminatory impacts and unfair biases that are prohibited by Union or national law.
• Article 10 sets out data and data governance requirements for high-risk AI systems and includes a requirement to examine and assess possible bias in training, validation and testing data sets.
• Deployers are required to ensure that any input data is relevant and sufficiently representative in view of the intended purpose of the high-risk AI system (Article 26(4)).

The Framework addresses the issue of bias (most notably in paragraphs 27-37 relating to ‘Non-bias and non-discrimination') and highlighted that AI has the potential to create and reinforce biases and that bias and discrimination by AI can cause manifest harm to individuals and to society. The European Parliament stated that regulation should encourage the development and sharing of strategies to counter these risks, including debiasing datasets in research and development and by the development of rules on data processing. The European Parliament also considered this approach to have the potential to turn software, algorithms and data into an asset in fighting bias and discrimination in certain situations, and a force for equal rights and positive social change.

At its core, the EU AI Act is driven by the imperative to safeguard the fundamental rights of EU citizens. The rapid advancement of AI technologies has introduced significant benefits but also potential risks, such as biases in decision-making systems and privacy infringements. The AI Act aims to mitigate these risks by establishing clear rules that ensure AI systems respect the rights enshrined in the EU Charter of Fundamental Rights. This focus on human-centric AI seeks to enhance trust and acceptance among the public, thereby promoting wider adoption of AI technologies in a responsible manner.

Within the EU AI Act, non-discrimination and fairness is incorporated withing the following:

• Recital 27 includes seven principles for trustworthy AI including ensuring that AI systems are developed and used in a way that includes diverse actors and promotes equal access, gender equality and cultural diversity, while avoiding discriminatory impacts and unfair biases that are prohibited by Union or national law.
• Article 10 sets out data and data governance requirements for high-risk AI systems and includes a requirement to examine and assess possible bias in training, validation and testing data sets.
• Deployers are required to ensure that any input data is relevant and sufficiently representative in view of the intended purpose of the high-risk AI system (Article 26(4)).

The Framework addresses the issue of bias (most notably in paragraphs 27-37 relating to ‘Non-bias and non-discrimination') and highlighted that AI has the potential to create and reinforce biases and that bias and discrimination by AI can cause manifest harm to individuals and to society. The European Parliament stated that regulation should encourage the development and sharing of strategies to counter these risks, including debiasing datasets in research and development and by the development of rules on data processing. The European Parliament also considered this approach to have the potential to turn software, algorithms and data into an asset in fighting bias and discrimination in certain situations, and a force for equal rights and positive social change.

Laws specifically addressing AI have not been introduced in Mauritius yet. However, the Data Protection Act 2017 provides that every controller or processor shall ensure that personal data are processed lawfully and fairly in relation to any data subject. The Blueprint aims at promoting digital inclusion, ensuring equitable access to services, and upholding human rights including accessibility, data privacy and non-discrimination.

Laws specifically addressing AI have not been introduced in Mexico yet. Article 2 of the AI Bill states as follows:

‘In any use of artificial intelligence systems, the protection of human rights must be guaranteed, and therefore any form of discrimination based on ethnic or national origin, gender, age, disabilities, social status, health conditions, religion, opinions, sexual preferences, marital status or any other practice that violates human dignity and aims to, or results in, nullifying or impairing the rights and freedoms of individuals is prohibited in their development and use.’

Laws specifically addressing AI have not been introduced in Morocco yet, so there are no AI-specific fairness or unlawful bias requirements. However, Morocco's existing constitutional and legal framework, including provisions of the Moroccan Constitution of 2011 guaranteeing equality and non-discrimination, and Law No. 09-08 on data protection, may apply to the extent that AI systems produce discriminatory outcomes or involve the unfair processing of personal data.

At its core, the EU AI Act is driven by the imperative to safeguard the fundamental rights of EU citizens. The rapid advancement of AI technologies has introduced significant benefits but also potential risks, such as biases in decision-making systems and privacy infringements. The AI Act aims to mitigate these risks by establishing clear rules that ensure AI systems respect the rights enshrined in the EU Charter of Fundamental Rights. This focus on human-centric AI seeks to enhance trust and acceptance among the public, thereby promoting wider adoption of AI technologies in a responsible manner.

Within the EU AI Act, non-discrimination and fairness is incorporated withing the following:

• Recital 27 includes seven principles for trustworthy AI including ensuring that AI systems are developed and used in a way that includes diverse actors and promotes equal access, gender equality and cultural diversity, while avoiding discriminatory impacts and unfair biases that are prohibited by Union or national law.
• Article 10 sets out data and data governance requirements for high-risk AI systems and includes a requirement to examine and assess possible bias in training, validation and testing data sets.
• Deployers are required to ensure that any input data is relevant and sufficiently representative in view of the intended purpose of the high-risk AI system (Article 26(4)).

The Framework addresses the issue of bias (most notably in paragraphs 27-37 relating to ‘Non-bias and non-discrimination') and highlighted that AI has the potential to create and reinforce biases and that bias and discrimination by AI can cause manifest harm to individuals and to society. The European Parliament stated that regulation should encourage the development and sharing of strategies to counter these risks, including debiasing datasets in research and development and by the development of rules on data processing. The European Parliament also considered this approach to have the potential to turn software, algorithms and data into an asset in fighting bias and discrimination in certain situations, and a force for equal rights and positive social change.

Laws specifically addressing AI have not been introduced in New Zealand yet, so there are no AI-specific fairness or unlawful bias requirements. Fairness and unlawful bias requirements under existing legislation could be applied in the AI context, such as the Human Rights Act 1993 (Human Rights Act). While it does not specifically regulate AI, the Human Rights Act is to be read as applying as widely as possible to protect human rights. Therefore, if an AI decision is ultimately attributable to a company or public body, the Human Rights Act would apply to that decision and create an obligation to ensure that decision is not discriminatory. The Biometrics Code reinforces this by prohibiting the use of biometric systems to categorise individuals into classes corresponding to the prohibited discrimination grounds in section 21(1) of the Human Rights Act, except in limited circumstances. Additionally, the OPC AI Guidance and AI Guidance for Business both emphasise fairness and bias mitigation as key considerations for responsible AI deployment.

A standalone AI law has not yet been enacted in Nigeria. However, other laws applicable to AI such as the Nigerian Constitution 1999 (as amended) and the Nigeria Data Protection Act, 2023 (NDPA) contain provisions addressing fairness and discrimination. The NDPA also includes provisions on automated decision-making that may be relevant to AI systems processing personal data.

The content on Fairness / unlawful bias in the European Union applies in Norway.

The AI Regulation expressly addresses fairness and unlawful bias through its principles on non-discrimination, transparency, ethics and accountability. It defines algorithmic bias and requires measures to prevent, mitigate and correct discriminatory or biased outcomes in the development, implementation and use of AI systems.

Certain discriminatory uses are expressly prohibited, including the use of biometric data to infer sensitive characteristics or to classify individuals or groups in a manner that generates discriminatory or disproportionate outcomes. It also classifies as high-risk any AI use that involves a high probability of stigmatisation, discrimination or cultural bias.

In addition, the AI Regulation requires public entities to conduct security audits to identify and correct vulnerabilities and biases in AI systems. It also requires developers, implementers and users to act ethically, avoid unfair or discriminatory outcomes, and implement best practices, standards or technical rules to identify and minimise bias in algorithms and datasets.

At its core, the EU AI Act is driven by the imperative to safeguard the fundamental rights of EU citizens. The rapid advancement of AI technologies has introduced significant benefits but also potential risks, such as biases in decision-making systems and privacy infringements. The AI Act aims to mitigate these risks by establishing clear rules that ensure AI systems respect the rights enshrined in the EU Charter of Fundamental Rights. This focus on human-centric AI seeks to enhance trust and acceptance among the public, thereby promoting wider adoption of AI technologies in a responsible manner.

Within the EU AI Act, non-discrimination and fairness is incorporated withing the following:

• Recital 27 includes seven principles for trustworthy AI including ensuring that AI systems are developed and used in a way that includes diverse actors and promotes equal access, gender equality and cultural diversity, while avoiding discriminatory impacts and unfair biases that are prohibited by Union or national law.
• Article 10 sets out data and data governance requirements for high-risk AI systems and includes a requirement to examine and assess possible bias in training, validation and testing data sets.
• Deployers are required to ensure that any input data is relevant and sufficiently representative in view of the intended purpose of the high-risk AI system (Article 26(4)).

The Framework addresses the issue of bias (most notably in paragraphs 27-37 relating to ‘Non-bias and non-discrimination') and highlighted that AI has the potential to create and reinforce biases and that bias and discrimination by AI can cause manifest harm to individuals and to society. The European Parliament stated that regulation should encourage the development and sharing of strategies to counter these risks, including debiasing datasets in research and development and by the development of rules on data processing. The European Parliament also considered this approach to have the potential to turn software, algorithms and data into an asset in fighting bias and discrimination in certain situations, and a force for equal rights and positive social change.

At its core, the EU AI Act is driven by the imperative to safeguard the fundamental rights of EU citizens. The rapid advancement of AI technologies has introduced significant benefits but also potential risks, such as biases in decision-making systems and privacy infringements. The AI Act aims to mitigate these risks by establishing clear rules that ensure AI systems respect the rights enshrined in the EU Charter of Fundamental Rights. This focus on human-centric AI seeks to enhance trust and acceptance among the public, thereby promoting wider adoption of AI technologies in a responsible manner.

Within the EU AI Act, non-discrimination and fairness is incorporated withing the following:

• Recital 27 includes seven principles for trustworthy AI including ensuring that AI systems are developed and used in a way that includes diverse actors and promotes equal access, gender equality and cultural diversity, while avoiding discriminatory impacts and unfair biases that are prohibited by Union or national law.
• Article 10 sets out data and data governance requirements for high-risk AI systems and includes a requirement to examine and assess possible bias in training, validation and testing data sets.
• Deployers are required to ensure that any input data is relevant and sufficiently representative in view of the intended purpose of the high-risk AI system (Article 26(4)).

The Framework addresses the issue of bias (most notably in paragraphs 27-37 relating to ‘Non-bias and non-discrimination') and highlighted that AI has the potential to create and reinforce biases and that bias and discrimination by AI can cause manifest harm to individuals and to society. The European Parliament stated that regulation should encourage the development and sharing of strategies to counter these risks, including debiasing datasets in research and development and by the development of rules on data processing. The European Parliament also considered this approach to have the potential to turn software, algorithms and data into an asset in fighting bias and discrimination in certain situations, and a force for equal rights and positive social change.

At its core, the EU AI Act is driven by the imperative to safeguard the fundamental rights of EU citizens. The rapid advancement of AI technologies has introduced significant benefits but also potential risks, such as biases in decision-making systems and privacy infringements. The AI Act aims to mitigate these risks by establishing clear rules that ensure AI systems respect the rights enshrined in the EU Charter of Fundamental Rights. This focus on human-centric AI seeks to enhance trust and acceptance among the public, thereby promoting wider adoption of AI technologies in a responsible manner.

Within the EU AI Act, non-discrimination and fairness is incorporated withing the following:

• Recital 27 includes seven principles for trustworthy AI including ensuring that AI systems are developed and used in a way that includes diverse actors and promotes equal access, gender equality and cultural diversity, while avoiding discriminatory impacts and unfair biases that are prohibited by Union or national law.
• Article 10 sets out data and data governance requirements for high-risk AI systems and includes a requirement to examine and assess possible bias in training, validation and testing data sets.
• Deployers are required to ensure that any input data is relevant and sufficiently representative in view of the intended purpose of the high-risk AI system (Article 26(4)).

The Framework addresses the issue of bias (most notably in paragraphs 27-37 relating to ‘Non-bias and non-discrimination') and highlighted that AI has the potential to create and reinforce biases and that bias and discrimination by AI can cause manifest harm to individuals and to society. The European Parliament stated that regulation should encourage the development and sharing of strategies to counter these risks, including debiasing datasets in research and development and by the development of rules on data processing. The European Parliament also considered this approach to have the potential to turn software, algorithms and data into an asset in fighting bias and discrimination in certain situations, and a force for equal rights and positive social change.

There is no AI‑specific fairness or anti‑bias statute.

SDAIA guidance emphasises fairness and encourages measures to eliminate data set bias and ensure inclusivity.

Laws specifically addressing AI have not yet been introduced in Singapore.

Fairness constitutes one of the guiding principles in the Model Framework. More specifically, it recommends:

• ensuring that algorithmic decisions do not create discriminatory or unjust impact across different demographic lines (e.g. race, sex, etc.);
• developing and including monitoring and accounting mechanisms to avoid unintentional discrimination when implementing decision-making systems; and
• consulting a diversity of voices and demographics when developing systems, applications and algorithms.

The Model Framework for Agentic AI highlights biased or unfair actions as one of the key risks associated with agentic AI.

The Principles recommend the following fairness and ethics practices:

• individuals or groups of individuals must not be systematically disadvantaged through AIDA-driven decisions unless these decisions can be justified;
• use of personal attributes as input factors for AIDA-driven decisions is justified;
• data and models used for AIDA-driven decisions must be regularly reviewed and validated for accuracy and relevance, and to minimise unintentional bias;
• AIDA-driven decisions must be regularly reviewed so that models behave as designed and intended;
• use of AIDA is aligned with the firm’s ethical standards, values and codes of conduct; and
• AIDA-driven decisions are held to at least the same ethical standards as human-driven decisions.

At its core, the EU AI Act is driven by the imperative to safeguard the fundamental rights of EU citizens. The rapid advancement of AI technologies has introduced significant benefits but also potential risks, such as biases in decision-making systems and privacy infringements. The AI Act aims to mitigate these risks by establishing clear rules that ensure AI systems respect the rights enshrined in the EU Charter of Fundamental Rights. This focus on human-centric AI seeks to enhance trust and acceptance among the public, thereby promoting wider adoption of AI technologies in a responsible manner.

Within the EU AI Act, non-discrimination and fairness is incorporated withing the following:

• Recital 27 includes seven principles for trustworthy AI including ensuring that AI systems are developed and used in a way that includes diverse actors and promotes equal access, gender equality and cultural diversity, while avoiding discriminatory impacts and unfair biases that are prohibited by Union or national law.
• Article 10 sets out data and data governance requirements for high-risk AI systems and includes a requirement to examine and assess possible bias in training, validation and testing data sets.
• Deployers are required to ensure that any input data is relevant and sufficiently representative in view of the intended purpose of the high-risk AI system (Article 26(4)).

The Framework addresses the issue of bias (most notably in paragraphs 27-37 relating to ‘Non-bias and non-discrimination') and highlighted that AI has the potential to create and reinforce biases and that bias and discrimination by AI can cause manifest harm to individuals and to society. The European Parliament stated that regulation should encourage the development and sharing of strategies to counter these risks, including debiasing datasets in research and development and by the development of rules on data processing. The European Parliament also considered this approach to have the potential to turn software, algorithms and data into an asset in fighting bias and discrimination in certain situations, and a force for equal rights and positive social change.

At its core, the EU AI Act is driven by the imperative to safeguard the fundamental rights of EU citizens. The rapid advancement of AI technologies has introduced significant benefits but also potential risks, such as biases in decision-making systems and privacy infringements. The AI Act aims to mitigate these risks by establishing clear rules that ensure AI systems respect the rights enshrined in the EU Charter of Fundamental Rights. This focus on human-centric AI seeks to enhance trust and acceptance among the public, thereby promoting wider adoption of AI technologies in a responsible manner.

Within the EU AI Act, non-discrimination and fairness is incorporated withing the following:

• Recital 27 includes seven principles for trustworthy AI including ensuring that AI systems are developed and used in a way that includes diverse actors and promotes equal access, gender equality and cultural diversity, while avoiding discriminatory impacts and unfair biases that are prohibited by Union or national law.
• Article 10 sets out data and data governance requirements for high-risk AI systems and includes a requirement to examine and assess possible bias in training, validation and testing data sets.
• Deployers are required to ensure that any input data is relevant and sufficiently representative in view of the intended purpose of the high-risk AI system (Article 26(4)).

The Framework addresses the issue of bias (most notably in paragraphs 27-37 relating to ‘Non-bias and non-discrimination') and highlighted that AI has the potential to create and reinforce biases and that bias and discrimination by AI can cause manifest harm to individuals and to society. The European Parliament stated that regulation should encourage the development and sharing of strategies to counter these risks, including debiasing datasets in research and development and by the development of rules on data processing. The European Parliament also considered this approach to have the potential to turn software, algorithms and data into an asset in fighting bias and discrimination in certain situations, and a force for equal rights and positive social change.

Laws specifically addressing AI have not been introduced in South Africa, so there are no AI-specific fairness or unlawful bias requirements.

Currently, the AI Act does not clearly stipulate this, but it is recommended in the National Guidelines for AI Ethics and other similar documents referred to in the Regulatory Guidance / Voluntary Codes section.

At its core, the EU AI Act is driven by the imperative to safeguard the fundamental rights of EU citizens. The rapid advancement of AI technologies has introduced significant benefits but also potential risks, such as biases in decision-making systems and privacy infringements. The AI Act aims to mitigate these risks by establishing clear rules that ensure AI systems respect the rights enshrined in the EU Charter of Fundamental Rights. This focus on human-centric AI seeks to enhance trust and acceptance among the public, thereby promoting wider adoption of AI technologies in a responsible manner.

Within the EU AI Act, non-discrimination and fairness is incorporated withing the following:

• Recital 27 includes seven principles for trustworthy AI including ensuring that AI systems are developed and used in a way that includes diverse actors and promotes equal access, gender equality and cultural diversity, while avoiding discriminatory impacts and unfair biases that are prohibited by Union or national law.
• Article 10 sets out data and data governance requirements for high-risk AI systems and includes a requirement to examine and assess possible bias in training, validation and testing data sets.
• Deployers are required to ensure that any input data is relevant and sufficiently representative in view of the intended purpose of the high-risk AI system (Article 26(4)).

The Framework addresses the issue of bias (most notably in paragraphs 27-37 relating to ‘Non-bias and non-discrimination') and highlighted that AI has the potential to create and reinforce biases and that bias and discrimination by AI can cause manifest harm to individuals and to society. The European Parliament stated that regulation should encourage the development and sharing of strategies to counter these risks, including debiasing datasets in research and development and by the development of rules on data processing. The European Parliament also considered this approach to have the potential to turn software, algorithms and data into an asset in fighting bias and discrimination in certain situations, and a force for equal rights and positive social change.

At its core, the EU AI Act is driven by the imperative to safeguard the fundamental rights of EU citizens. The rapid advancement of AI technologies has introduced significant benefits but also potential risks, such as biases in decision-making systems and privacy infringements. The AI Act aims to mitigate these risks by establishing clear rules that ensure AI systems respect the rights enshrined in the EU Charter of Fundamental Rights. This focus on human-centric AI seeks to enhance trust and acceptance among the public, thereby promoting wider adoption of AI technologies in a responsible manner.

Within the EU AI Act, non-discrimination and fairness is incorporated withing the following:

• Recital 27 includes seven principles for trustworthy AI including ensuring that AI systems are developed and used in a way that includes diverse actors and promotes equal access, gender equality and cultural diversity, while avoiding discriminatory impacts and unfair biases that are prohibited by Union or national law.
• Article 10 sets out data and data governance requirements for high-risk AI systems and includes a requirement to examine and assess possible bias in training, validation and testing data sets.
• Deployers are required to ensure that any input data is relevant and sufficiently representative in view of the intended purpose of the high-risk AI system (Article 26(4)).

The Framework addresses the issue of bias (most notably in paragraphs 27-37 relating to ‘Non-bias and non-discrimination') and highlighted that AI has the potential to create and reinforce biases and that bias and discrimination by AI can cause manifest harm to individuals and to society. The European Parliament stated that regulation should encourage the development and sharing of strategies to counter these risks, including debiasing datasets in research and development and by the development of rules on data processing. The European Parliament also considered this approach to have the potential to turn software, algorithms and data into an asset in fighting bias and discrimination in certain situations, and a force for equal rights and positive social change.

There is currently no specific concept of unlawful bias or fairness in relation to AI prescribed under Thai law. However, the Draft AI Law Principles (2025) broadly set out the principles relating to the fundamental concepts of AI (e.g. the principle of non-discrimination in relation to actions arising from AI, principle recognising AI as merely a tool of humans, and principle concerning exceptions for acts arising from AI that could not have been reasonably foreseen). The details remain to be finalised.

Laws specifically addressing AI have not been enacted in Türkiye yet. NAIS sets out an ‘AI Principle’ of ‘Fairness’, as follows (page 60 of NAIS):

"AI systems should be designed to provide an equal and fair service to all stakeholders while adhering to the rule of law and fundamental rights and freedoms. The fairness of AI systems means that the benefits of AI technology are shared at local, national and international levels, while taking into account the specific needs of different age groups, different cultural systems, different language groups, people with disabilities, and disadvantaged, marginalized and vulnerable segments of the society. It should be ensured that decisions made based on algorithms do not give rise to discriminatory or unfair effects on different demographic populations. In order to prevent the emergence of unintentional discrimination in decision-making processes, monitoring and accountability mechanisms should be developed and those mechanisms should be included in the implementation process."

There is no unified federal law or emirate level law in the UAE that has a primary focus on regulating AI (and therefore no binding obligations in relation to fairness and bias).

However, the AI Ethics Guide contains a principle of fairness which provides that:

• Data ingested should, where possible, be accurate and representative of the population.

• Algorithms should avoid non-operational bias.

• Steps should be taken to mitigate and disclose the biases inherent in datasets.

• Significant decisions should be provably fair.

• All personnel involved in the development, deployment and use of AI Systems have a role and responsibility to operationalize AI fairness and should be educated accordingly.

The DIFC’s Data Protection Regulations also provide that AI Systems must be designed in accordance with the principle of fairness. In particular, AI Systems should be designed to treat all individuals equally and fairly, regardless of race, gender, or other specifically subjective factors; and AI Systems should be designed to avoid potential biases, including unjust bias, or where possible, mitigate bias that could lead to unfair outcomes.

The principle of fairness identified in the White Paper specifies that AI systems should not undermine the legal rights of individuals or organisations, discriminate unfairly against individuals or create unfair market outcomes. Since AI can have a significant impact on people’s lives, the principle states that AI-enabled decisions with high impact outcomes should not be arbitrary and should be justifiable. Deployment of AI systems with specific biases could breach existing laws, including the Equality Act 2010, the Data Protection Act 2018 and/or various employment laws, depending on context.  

As with transparency, there is no federal law in the US that specifically addresses fairness, bias, or other forms of algorithmic discrimination in AI systems. Under the Biden Administration, federal agencies sought to address these issues by applying existing civil rights, employment, and consumer protection laws to AI use cases. However, this activity has almost entirely ended, and agency-issued guidance on these subjects has in some cases been removed from public websites. Meanwhile, however, several states have enacted or proposed legislation to directly address algorithmic discrimination. For example:

• California’s Fair Employment and Housing Act applies to employers’ use of ‘[AI], algorithms, and other automated-decision systems’ in employment decisions
• Colorado’s new automated decision-making technology law, which splits liability for algorithmic discrimination between developers and deployers under the Colorado Anti-Discrimination Act—holding each responsible only to the extent of its relative fault, depending on whether a Covered ADMT was used as intended and documented—and voids contractual indemnification clauses that attempt to shift liability for unlawful algorithmic discrimination in consequential decisions
• Illinois has enacted workplace AI legislation that prohibits the use of AI in hiring or employment decisions that could result in discrimination
• New Jersey issued guidance clarifying that the New Jersey Law Against Discrimination (LAD) applies to “algorithmic discrimination” resulting from the use of AI and other decision-making tools, including in employment
• New York City’s Local Law 144 requires annual bias audits for automated employment decision tools and mandates candidate notification

These state and local efforts, combined with prior federal activity, may reflect a growing—though not entirely shared—belief that AI systems can perpetuate or amplify existing societal biases, and that legal frameworks are evolving to ensure fairness, particularly in domains like employment, housing, and healthcare.