Introduction
As artificial intelligence (AI) and automated decision-making (ADM) systems become increasingly integrated into social life, ethical considerations surrounding their use are more crucial than ever. A recent study sheds light on how disabled individuals form ethical judgments about these technologies, highlighting three central themes: embodiment and relationality, recognition, and justice.
Key Themes in Ethical Judgments
1. Embodiment and Relationality
Participants emphasized the vital importance of human connection, care, and empathy in decision-making processes, especially in health and social care contexts. They expressed concerns that ADM systems cannot replicate the nuanced human interactions essential for fair and just outcomes. However, some participants recognized potential benefits in utilizing technology to support human interactions, provided it does not entirely replace them.
2. Recognition
The recognition of disability’s complexities was a significant concern for participants. Many shared experiences of misrecognition in various systems, underscoring that well-designed ADM systems should accommodate and value the diverse experiences of disabled individuals. Conversely, poorly designed systems could exacerbate issues of misrecognition and lead to injustices.
3. Justice
Participants frequently discussed the broader political and economic implications of ADM systems. They viewed these technologies as potential tools for perpetuating existing inequalities, often prioritizing efficiency and profit over fairness. The need for individual autonomy and involvement in decision-making processes was especially emphasized in high-stakes situations.
Balancing Human and Machine Decision-Making
The study revealed mixed views on the balance between human and machine decision-making:
- Human Interaction and Empathy: Many participants stressed the importance of human connection and empathy. They believed that human decision-makers, despite their flaws, could provide the nuanced understanding necessary for fair outcomes.
- Limitations of Machines: There was a general consensus that ADM systems lack the ability to replicate the depth of human interactions, particularly the emotional and physical dimensions of care.
- Potential Benefits of Automation: Some participants acknowledged that technology could support human decision-making, offering impartiality and consistency that might be absent in human assessments.
- Cautious Endorsement: Participants expressed cautious support for technology, provided it enhanced rather than diminished human relationships. They stressed that automation should complement human judgment.
- Concerns About Full Automation: There was significant wariness regarding fully automated decision-making in critical situations, as this could undermine individual autonomy and lead to feelings of devaluation.
Conclusion
The findings from this study emphasize that AI ethics must transcend mere technical functionality to address the broader social and political dimensions of technology. Ethical frameworks should authentically represent the diversity of disability experiences and consider the specific social relationships and contexts in which technology operates. Ultimately, a nuanced approach to AI ethics is necessary—one that prioritizes human flourishing and justice, ensuring that disabled voices are central to the discourse on technology’s role in society.
For further insights, you can read the full study here.
Source: van Toorn, G., Scully, J.L. & Gendera, S. “This robot is dictating her next steps in life”: disability justice and relational AI ethics. AI & Soc(2025). https://doi.org/10.1007/s00146-025-02224-x
