Practical ethics for the AI age
AI is here to stay, and it is affecting the way we think and behave, while also raising many challenging questions. Should doctors trust AI-generated diagnoses? Should hiring and firing decisions be delegated to AI managers? Could family disputes be resolved through AI mediation? The EU-funded AIOLIA(opens in new window) project was launched in 2025 to tackle these issues by developing AI ethics guidelines, designing instructional materials and conducting large-scale training in AI ethics. Going beyond the question of whether AI systems are ethical, the project explores how they affect our judgement and behaviour, and what safeguards need to be set in response. “The starting premise of AIOLIA is about ‘what is the role of technology for human beings?’,” states lead researcher Petra Saskia Bayerl in a recent news item(opens in new window). Bayerl – who is a professor at the Centre of Excellence in Terrorism, Resilience, Intelligence and Organised Crime Research, a research centre at AIOLIA project partner Sheffield Hallam University in the United Kingdom – argues that AI’s ability to shape our behaviour and cognition is what makes it fundamentally different from other technologies. “AI is different because it increasingly takes over tasks we traditionally associated with human thinking. It can make cognitive work more efficient, automate certain decisions and identify patterns that we once considered uniquely human capabilities,” she explains.
AI ethics in Europe and beyond
It is these concerns that led AIOLIA to release concrete AI ethics guidelines(opens in new window) for six European use cases, co-created by leading academic, policy and ethics-aware industrial partners from diverse sectors. Four use cases focused on changes in human expertise and professional behaviour when AI is used by: doctors for diagnosing and treating patients; safety engineers to speed up software release approvals; recruiters when hiring workers; and security professionals for detecting hate speech. A final two use cases – involving AI systems used as individuals’ and families’ virtual assistants, and deepfake therapy for processing trauma and grief – concerned changes in human cognition and private behaviour. This work was extended through four international use cases(opens in new window) that explored the implementation of ethical principles across different national contexts. The use cases covered AI systems as grief-support personal assistants (Canada), AI systems for smart elderly care (China), workplaces equipped with AI tools for behavioural analysis (Japan) and AI systems as personal companions to assist senior citizens (South Korea). Based on the project’s findings, ethical concerns depend largely on context. In the professional use cases, the ethical requirements emphasised were accountability, transparency and non-bias. In the private ones, the focus was more on individual safety, well-being, user autonomy and agency. More recently, AIOLIA developed a tool for evaluating ethics readiness levels and algorithmic impact(opens in new window). It helps ethics experts and technical experts conduct semi-structured dialogues that can inspire improved AI system design. “AIOLIA is ultimately about making ethics practical,” Bayerl states. The safeguards proposed by the AIOLIA (OPERATIONALIZING AI ETHICS FOR LEARNING AND PRACTICE: A GLOBAL APPROACH) project, including human oversight, appeals mechanisms, staff training and regular policy reviews, ultimately all have one goal: to make AI ethics work in everyday life. For more information, please see: AIOLIA project website(opens in new window)
