Human Rights and AI - Protecting People

Artificial Intelligence is reshaping the world. From healthcare diagnostics to online platforms, its impact is felt across everyday life.

Artificial Intelligence is reshaping the world. From healthcare diagnostics to online platforms, its impact is felt across everyday life. However, alongside opportunity, AI comes with risks. AI can strengthen human rights, or it can undermine them. The difference lies in how these technologies are designed, governed, and deployed.

As UNESCO puts it:

“The protection of human rights and dignity is the cornerstone of the [AI Ethics] Recommendation . . . always remembering that the importance of human oversight of AI systems”. (UNESCO, Ethics of Artificial Intelligence)

UNESCO explores the relationship between Ai and human rights, the different perspectives at play, real-world examples (good, bad, and thought provoking), and the guidance available from international bodies.

Different views of AI and Human Rights

Opportunity View

AI can help people realise their rights. Tools that detect disease early support the right to health, accessibility focused AI (such as speech-to-text or language translation) enhances the right to education and communication; and content moderation systems can reduce online harassment, upholding the right to dignity.

Risk View

Poorly designed AI can harm rights, Biased hiring algorithms may deny equal opportunities, predictive policing has targeted minority groups unfairly, and large-scale surveillance threatens privacy and freedom of expression.

Governance View

AI itself is not inherently good or bad. What matters is governance, transparency, and accountability. Rights must be safeguarded through regulation, design choices, and effective oversight.

How does AI intersect with Human Rights?

Ai interacts with many human rights in different ways. Some of the most important rights include:

Privacy: AI often relies on vast amounts of personal data. When data is collected or used without proper consent, it raises concerns about surveillance and intrusion. For example, facial recognition technology has parked global debates about privacy. String data protection laws and safeguards such as anonymisation and purpose limitation are key to protecting this right.
Equality and non-discrimination: If training data reflects existing social biases, AI systems can reproduce and even amplify unfairness. In hiring, for instance, algorithms have been shown to disadvantage women and minority groups. Tackling this requires fairness audits, diverse datasets, and ongoing monitoring.
Freedom of expression: Automated systems that moderate online content can help reduce harmful speech, but they can also mistakenly silence voices. Particularly those from marginalised groups. Transparency in moderation policies and accessible appeals processes are essential to balance safety with free expression.
Access to benefits and opportunities: AI has the potential to widen the digital divide if only certain groups or regions benefit from it. For example, AI tools in educations or healthcare may be accessible n wealthy areas but absent elsewhere, leading to inequality. Designing systems inclusively and ensuring broad access helps prevent this gap from growing.
Due process and remedy: When AI is used in decisions about people’s lives, such as welfare, loans, or legal cases, the risk of opaque “blackbox” decisions is high. If people cannot understand or challenge the outcome, their rights to due process and remedy are undermined. To address this, systems must offer explanations, audit trails, and ways to appeal or contest decisions.

The United Nations’ view is clear:

“AI actors [must] make all reasonable efforts to minimise and avoid reinforcing or perpetuating . . .discriminatory or biased outcomes throughout the lifecycle of the AI system.” (United Nations, Artificial Intelligence and Creativity)

Let’s look at some case studies in practice . . .

Harmful: Dutch Childcare Benefits Scandal: In the Netherlands, thousand of parent were wrongly accused or childcare benefits fraud after algorithms flagged them as “high-risk”. Many were forced into debt. Lost homes, or event had children removed from their care. Ethnic minorities were disproportionately targeted due to biased profiling. A parliamentary inquiry found that the rule of law had been violated, and the Dutch government resigned in 2021 as a result.

Positive: AI in Healthcare: In the United Kingdom, AI is being used to reduce NHS diagnostic backlogs. Radiology Ai systems can help detect cancers, improving patient outcomes and supporting the right to health. With proper oversight and humans-in-the-loop, healthcare AI demonstrates how technology can advance when aligned with ethics.

Interesting: Estonia’s “AI Judge”: Estonia has piloted an AI system to help resolve small claims disputes, The goal is to make justice faster and cheaper. However, the question remains: How can citizens appeal? Who is accountable for errors? The experiment shows the tension between efficiency and the right to a fair trial.

Additional guidance and standards from trusted bodies

United Nations Guiding Principles on Business and Human Rights sets out global expectations for companies to respect human rights, including in AI. (UN, OHCHR and business and human rights)
OECD AI principles emphasise fairness, transparency, and accountability, adopted by 46 countries. (OCED, Principles for trustworthy AI)
UNESCO Recommendation on the Ethics of AI (2021) was adopted by 193 countries, making human dignity and rights the foundation for global AI ethics. (UNESCO, Ethics of Artificial Intelligence)

Why does this matter in Northern Ireland?

In Northern Ireland, where equality and fairness are cornerstones of governance, the human rights dimension of AI cannot be ignored. Lessons from global failures like the Dutch benefits scandal show that harm is not hypothetical, it is real, systemic, and avoidable.

By aligning with established frameworks, businesses and policymakers in Northern Ireland can:

Conduct Human Rights Impacts Assessment (HRIA) before deploying AI
Embed fairness and transparency into design
Guarantee contestability and remedy where AI is used in decisions about people’s lives
Engagement communities and stakeholders in shaping AI governance

Final Thoughts

Ai and human rights are inseparable. When technology ignores rights, people suffer. When, it upholds them, society benefits. The choice belongs to people.

Building responsible AI means not just chasing innovation, but ensuring that every step respects dignity, fairness, and equality. In the words of the Council of Europe, AI governance must “protect human rights, democracy, and the rule of law in the digital environment”. (Council of Europe and AI, Artificial intelligence and human rights)

Commitment and Disclaimer
The Responsible AI Hub provides resources for learning, examples of AI in action and support for responsible AI. Use of these materials does not create any legal obligations or liability with the AICC.

We'd love to hear from you

Innovation thrives through connection. Whether you're an SME, researcher, or professional exploring AI, we’re here to help.

Get In Touch

Go To AICC website

Go To Ulster University website

Go To Queens University Belfast Website

Go To Invest NI website

Go To DFE website

Our Story and Vision

Meet the Team

AICC in the News

AI Capability Census

Transformer Programme

Transformer Stories

Research and Development

Explore the Responsible AI Hub

Responsible AI Fundamentals

Responsible AI Tools

Responsible AI News & Insights

Qualifications and Skills Development

AI Acceleration Tour