SensAI: Advancing AI Ethics and Human Rights in Low-Resource Language Processing

The SensAI project is dedicated to bridging the gap between AI technology, human rights, and ethics, with a special focus on low-resource languages. SensAI builds upon the ALFIE project, expanding its ethical, legal, and technical dimensions to enhance the moral sensitivity of AI development and improve AI’s alignment with human rights principles. AI development must not only be technologically advanced but also transparent, fair, and accessible. By integrating legal, ethical, and AI research, SensAI aims to develop methodologies for assessing the impact of AI on human rights, improve the efficiency of large language models for Slovak and other low-resource languages, and enhance moral sensitivity frameworks for AI developers. SensAI is committed to making AI more inclusive, responsible, and aligned with fundamental human values.

A key focus of SensAI is ensuring that AI systems are ethical and accessible, particularly in the context of low-resource languages like Slovak. The project tackles three interconnected challenges: moral sensitivity in AI development, human rights impact assessment, and improving AI accessibility for underrepresented languages. By evaluating how AI practitioners recognise and address ethical dilemmas, extending Fundamental Rights Impact Assessments (FRIA) to AI applications, and enhancing large language models (LLMs) for Slovak, SensAI aims to develop methodologies that embed ethical considerations, transparency, and fairness directly into AI systems. This research is crucial for fostering trustworthy AI that aligns with societal values and regulatory frameworks.

Human Rights Impact Assessment for AI – The project extends Fundamental Rights Impact Assessments to analyse the ethical and legal implications of AI, particularly Automated Machine Learning (AutoML) systems, and ensures compliance with EU AI regulations.

Improving AI for Low-Resource Languages – SensAI enhances large language models (LLMs) for Slovak, making AI tools accessible for speakers of underrepresented languages while evaluating their performance against multilingual AI approaches.

Moral Sensitivity in AI Development – By studying ethical awareness in AI teams, SensAI provides frameworks that strengthen developers’ ability to identify and mitigate ethical risks in their work.

A key focus of SensAI is evaluating and strengthening moral sensitivity in AI development. This involves assessing how AI practitioners recognise, understand, and address ethical dilemmas in their work. By identifying measurable components of moral sensitivity, SensAI aims to develop practical methodologies for long-term monitoring, ensuring that ethical considerations are not just theoretical discussions but are effectively embedded into AI systems. This research is essential for fostering trustworthy AI that respects fundamental human rights and adapts to evolving societal and regulatory landscapes.

SensAI represents a forward-thinking approach to responsible AI, creating a future where AI technologies serve society in an equitable and just way. SensAI aims to bridge the gap between AI innovation and ethical responsibility, ensuring that AI systems are not only technically sophisticated but also aligned with human values. By fostering innovation while upholding human rights and ethical AI principles, SensAI is paving the way for a more inclusive and responsible AI ecosystem.