The use of AI in a variety of settings may raise human rights concerns. These include the potential of disproportionately harming one group of rights-holders in order to advance another; creating new forms of discrimination and inequality; or facilitating data-driven surveillance.
These questions raise challenging legal issues and gaps for international human rights law to address. This article outlines key considerations for States, laws and regulators and also ways AI can be used to contribute in International Human Rights Law.
How AI can be used to contribute to International Human Rights Law?
AI-driven insights are a powerful tool in international human rights law, helping to address and analyze complex issues related to human rights violations, accountability, and advocacy. Here are some of the ways AI can be used to contribute to this field.
Data Visualization and Analysis
AI is able to process large amounts of data, from different sources such as news articles, reports, and social media to identify patterns and trends related to human rights violations. The data can be visualized, which makes it easier for researchers, policymakers, and activists to communicate and understand the extent of the problem.
AI is a powerful tool that can be used to create early warning systems for detecting potential human rights violations before they escalate. AI algorithms are able to identify areas or situations with a high risk of human rights violations by analyzing real-time and historical information. This allows for proactive intervention.
AI is able to assist with the analysis of documents, court rulings, and treaties relating to international human rights law. It can assist researchers and legal professionals in quickly extracting relevant information, identifying precedents, as well as tracking legal developments over time.
International Human rights issues are often multilingual. AI-powered translation tools help organizations and advocates better understand and access information around the globe.
Risk and Sentiment Analysis
AI analyzes social media and news to gauge public sentiment and reactions to human rights issues. This can give insights into the public’s awareness, support, or backlash towards specific cases. These insights can be used to inform advocacy strategies.
AI is able to assess the risks associated with human rights violations such as ethnic cleansing, forced displacement, or torture. These risk assessments are useful for policymakers to allocate resources and prioritize interventions.
AI is an effective tool for risk assessment, which is crucial to preventing and addressing human rights violations. AI can assess risk factors for specific human rights violations, such as forced displacement, ethnic cleaning, and torture. AI systems, using advanced algorithms and data-analysis techniques, can analyze data from a variety of sources, including historical records, demographic data, geopolitical dynamics and social media trends to identify patterns and indicators.
For example, forced displacement is often a complex issue that involves factors like political instability, armed conflict, discrimination and socioeconomic disparities. AI can analyze all these variables and recognize correlations or anomalies that may indicate the possibility of displacement. AI can also process large volumes of data in cases of torture or ethnic cleansing to detect early warning signals, allowing authorities, humanitarian groups, and advocates to take proactive measures.
AI-driven risk assessments are able to provide timely, data-driven insights that can be used to inform policy decisions, allocation of resources, and prevention actions. This allows for more efficient resource allocation to high-risk areas and populations, which can lead to early interventions that could mitigate the severity of human rights violations. It is important to stress that AI-driven assessment of risk should be combined with human expertise and ethical considerations. AI can help us better understand and respond to human rights violations, but it must be done in a way that respects transparency, accountability, and the dignity of the individual. AI experts, human rights advocates, and policymakers should work together to harness AI’s potential for better protection and promotion of international human rights.
Predictive analytics
AI uses historical data to predict trends and future developments in human rights violations. This can help organizations and governments allocate resources and craft policies to effectively prevent or address such issues.
AI-powered legal research tools are able to help lawyers and academics in the area of international human rights law by quickly identifying cases, precedents, and legal arguments. This reduces the amount of time needed for legal research.
AI helps human rights organizations and activists to tailor their advocacy by analyzing preferences and behavior. This can result in more effective and targeted campaigns.
Ethical considerations
The use of artificial intelligence (AI) in various aspects of society, including government, business, and healthcare, raises significant ethical considerations with respect to human rights laws. While AI has the potential to bring about positive advancements and efficiencies, it also poses various risks and challenges to human rights. Here are some key ethical considerations:
- Privacy and Data Protection: AI systems often rely on vast amounts of personal data, raising concerns about privacy and data protection. Human rights laws, such as the General Data Protection Regulation (GDPR) in the European Union, seek to protect individuals’ rights to privacy and control over their personal information. The use of AI must comply with these laws by ensuring transparency, consent, and secure data handling.
- Bias and Discrimination: AI algorithms can perpetuate and even exacerbate biases present in training data. This can lead to discriminatory outcomes in areas like hiring, lending, and law enforcement. Ethical considerations require that AI systems are developed and monitored to prevent and mitigate bias, adhering to anti-discrimination laws and principles of fairness.
- Accountability and Transparency: Human rights laws emphasize the importance of accountability and transparency in decision-making processes. When AI systems make critical decisions that affect individuals’ rights (e.g., sentencing in criminal justice), it is essential to have mechanisms in place to explain how these decisions are reached and who is responsible. This is particularly important in ensuring due process and fair treatment.
- Access to AI Benefits: Ethical concerns arise when AI systems are not accessible to all segments of society, potentially exacerbating inequalities. Human rights laws stress the importance of equal access and non-discrimination. Ensuring that AI technologies are affordable, accessible, and usable by all, regardless of socioeconomic status, is crucial.
- Autonomy and Human Control: The deployment of AI in areas like autonomous weapons and surveillance can challenge the principle of human autonomy and control. Ethical considerations demand that human rights laws are upheld, emphasizing the importance of maintaining human agency over AI systems to prevent misuse and harm.
- Job Displacement: The automation of jobs through AI can have significant socioeconomic impacts, potentially violating the right to work and the right to an adequate standard of living. Governments and organizations using AI should consider the ethical implications and take measures to retrain and support individuals whose jobs are displaced by automation.
- Security and Discrimination: AI systems used in security and surveillance can infringe on individuals’ rights to privacy and freedom of expression. Striking a balance between security and civil liberties is a complex ethical challenge, requiring adherence to laws that protect these rights while addressing legitimate security concerns.
In conclusion, the ethical considerations surrounding the use of AI in the context of human rights laws are multifaceted. Striking the right balance between harnessing the benefits of AI and safeguarding individual rights is essential. This involves not only complying with existing human rights laws but also proactively addressing the unique ethical challenges that AI presents to ensure a fair and just society in the digital age. Regular review and adaptation of legal frameworks will be necessary to keep pace with AI’s evolving impact on human rights.
2. Artificial Intelligence and Human Rights
There has never been a more urgent time to invest in advancing human rights by developing a legal framework to govern AI, and ensuring that it does not amplify and reinforce existing social and economic inequalities. Regressive capitalists can use AI to comb worker social media posts and identify union organizers or ‘trouble makers’; corrupt governments can use facial recognition technology to target migrant workers or resurgent authoritarian leaders that they perceive as challenging employment practices and poor working conditions.
Moreover, if we look at the current state of global affairs – climate disasters, armed conflicts/assassinations/destabilization of democracies, mass displacement of people and the COVID-19 pandemic – it is clear that resurgent dictators and powerful capitalists are positioning themselves to style themselves as the true ‘authorities’ of a different version of international human rights. This makes it even more difficult to seek accountability for dictators and resurgent authoritarians that employ automation and AI tools in the service of human rights abuses, or dismiss those tools as ‘human rights imperialism’ or ‘Western-imposed human rights’.
The discussion of these legal issues using the frame of vulnerability will valuably help consolidate the identification of critical areas of concern and guide AI risk and impact mitigation efforts to better protect human and societal well-being.
The human rights dimensions of AI need to be central to discussions of governance and policy. This will require that the concerns of those who will be affected by AI technologies – such as women, minority groups and marginalized people – are given more prominence in these debates. It will also require attention to those sectors where a greater risk of bias in AI is likely to occur, such as in law enforcement, security and surveillance, justice, migration or financial services.
The scientific literature on AI & HRs highlights that there is considerable overlap in the areas affected by these technologies. This includes the right to life, the right to security and the right to privacy. Other important rights include the right to health, the right to non-discrimination and the right to work.
Despite this, the human rights aspects of AI are not fully considered in governance initiatives or by international bodies. The inclusion of human rights considerations in the European Commission’s new AI strategy – as well as in the Dutch bill of rights and the national AI Act – is an encouraging step, but more needs to be done. In particular, there must be a stronger culture of human rights impact assessments (HRIAs) in the development of AI technologies and the establishment of robust accountability mechanisms.
3. Human Rights and Machine Learning
The immense powers and capacities of AI raise existential questions and urgent challenges for human rights. They also pose risks that cannot be easily foreseen (as law and regulation often only emerge reactively to felt experiences and impacts of new technologies or fields of endeavor). This is why it is crucial for States, laws, and regulators to remember their constant obligation to protect fundamental human rights – even in the age of AI.
The systematic assessment and monitoring of the impact of AI on human rights is vital to ensure that its development, use, and operation respect international human rights standards. For instance, the biases in datasets used by AI tools can lead to discriminatory decisions – with acute risks for marginalized groups. This is why it is critical for companies and States to be transparent about the way they develop, manage, and apply AI systems.
It is a challenge that there is not enough attention to the systems nature of AI ecosystems. The discussion on the legal issues, gaps and challenges posed by AI is still too narrow and focused primarily on privacy and data protection. It is essential to reframe and strengthen the debate on AI and its human rights context, particularly through a focus on vulnerability. This will valuably help consolidate the identification of key areas of concern, guide AI risk and impact mitigation efforts and ensure that AI technologies advance human rights for everyone, especially those who are most vulnerable to its harmful effects.
4. Human Rights and Big Data
Despite the promise of AI, it may be used to perpetuate human rights abuses and threaten fundamental freedoms. As with other new technologies, it is a challenge for activists and governments alike to harness and use its potential for good while also addressing harm. Foundations that support human rights can establish technology funds and proactively invest in developing ethically sound technologies, and activists can work with tech companies to develop data analysis tools to monitor trends such as misogyny on Twitter.
At the same time, concerns about privacy and surveillance are not dissipated as governments acquire access to masses of data collected by tech giants and other private actors. Those anxieties could become even more pronounced as AI becomes more prevalent in the world, giving state and non-state actors alike an unprecedented ability to monitor and influence citizen lives, discriminate on the basis of a range of characteristics including gender and sexuality, and stifle dissent with impunity. As a result, it is vital that all stakeholders are equipped to understand and engage with the challenges associated with AI and big data in order to promote positive outcomes and prevent harm.