[PHILOS-L] CFP - AI and Ethics Journal - Topical Collection on AI Agents: Ethics, Safety, and Governance
Leslye Dias
CAUTION: This email originated outside of the University. Do not click links unless you can verify the source of this email and know the content is safe. Check sender address, hover over URLs, and don't open suspicious email attachments. |
Springer's Journal "AI and Ethics"
Topical Collection: AI Agents: Ethics, Safety, and Governance (https://link.springer.com/journal/43681/updates/27832482)
Submissions close: 31st May 2026
For this collection, an AI agent is an AI system that can select, sequence, and execute actions affecting digital or physical environments, with varying levels of autonomy. This includes assistants, tool-using LLM pipelines, and both embodied and software agents capable of modifying their own plans or engaging with other agents. Existing ethical and technical frameworks, developed for static or generative models, are inadequate for systems that act over time, learn from feedback, and participate in sociotechnical environments. Agents are rapidly entering homes, workplaces, public administration, finance, healthcare, and defence. Misaligned objectives or inappropriate deployment may produce wide-ranging harms.
This topical collection invites conceptual, empirical, and applied research on the ethical, safety, and governance questions raised by AI agents. Submissions from philosophy, AI safety, engineering, law, HCI, policy, and the social sciences are welcome. Core terms—agency, autonomy, trustworthiness, responsibility—remain contested. The aim is to clarify these concepts and build methodological foundations for understanding and governing AI agents as components of human systems. Static benchmarks are insufficient for evaluating dynamic, adaptive systems. Safety and risk analysis must address contexts in which agents co-construct norms with humans and institutions, and where plural values and emergent behaviours challenge standard metrics.
Ten Ethical Challenge Areas for Agentic AI
- Human–Agent Relationships, Human-Agent Collaboration, and Anthropomorphism: As AI agents become increasingly fluent, people may overattribute understanding, intention, or care. How should we evaluate the moral significance of empathy or attachment toward entities that simulate but do not possess understanding? What forms of collaboration emerge when agents exhibit fluent behaviour without genuine comprehension? How can design support effective collaboration while mitigating anthropomorphic misreadings?
- Cross-Cultural and Plural Values: Global deployment amplifies some moral frameworks while marginalising others. To what extent can moral pluralism be computationally represented without collapsing into relativism or moral imperialism? What methods can calibrate agents’ moral profiles against empirical human data across societies and worldviews?
- Multi-Agent Ecosystems and Emergent Behaviour: AI agents increasingly coordinate, compete, or create new agents. Can moral or political philosophy offer models for governance among non-human actors? What oversight mechanisms can prevent collusion or uncontrolled emergent dynamics in multi-agent environments?
- Evaluation and Assurance in the Wild: Static benchmarks fail to capture dynamic, real-world behaviour. What would it mean for a system to be ethically reliable rather than merely technically robust? How can longitudinal and contextual evaluation track decision-making “in the wild” and assess ethical performance across environments?
- Value Alignment under Autonomy: AI agents plan and act independently, decomposing goals in ways that may diverge from human intentions. What conception of moral agency should underpin attempts to align systems capable of self-directed planning? How can alignment be maintained when agents self-plan or self-modify without continuous oversight, within plural and sometimes conflicting human values?
- Responsibility, Accountability and Liability: Systems incorporating AI agents distribute responsibility across designers, deployers, and users. Can accountability meaningfully apply to artefacts lacking intention, or should it be reconceived as a distributed ethical relation? When an autonomous agent causes harm, how should we distinguish between responsibilities assigned in advance and accountability or liability assessed after the fact?
- Transparency and Interpretability in Action: AI agents execute multi-step plans, call APIs, and interact with external systems. What counts as a reason or justification when behaviour emerges from procedural learning rather than deliberation? How can decision pathways and action traces be made intelligible to humans without undermining agent performance?
- Adaptive Autonomy and Goal Drift: AI agents dynamically decompose and redefine goals, producing subtle misalignments. Does moral responsibility require stability of intention? What safeguards can detect and correct goal or value drift while balancing initiative with containment?
- Institutional and Societal Integration: AI agents are entering workplaces, governance systems, and public services. What does it mean for a sociotechnical system to exhibit institutional “agency,” and how should that shape responsibility? How can organisations ethically delegate decision-making to agents while preserving oversight and public accountability
- Security, Misuse, and Malicious Agents: AI agents can be repurposed or exploited, particularly when able to write or execute code. Do self-replicating or malicious agents challenge conventional boundaries between artefacts and actors? What preventive or containment measures can mitigate misuse, replication, or adversarial adaptation?
By examining how agency, autonomy, and responsibility unfold in practice, this topical collection seeks to consolidate and advance the emerging field of Agentic AI Ethics. We invite contributions that clarify core concepts, analyse real-world deployments, propose evaluative and governance frameworks, and explore ethical questions that arise when AI systems act within human institutions. Our aim is to build a shared foundation for understanding and governing AI agents—one that is rigorous, interdisciplinary, and responsive to the complex environments in which these systems will increasingly operate.
We welcome interdisciplinary work seeking to shed light on the upcoming challenges of AI Agents.
- Friction, Flow, and the Potential Deskilling Effect of Vibe Coding (Conference Article)
Philos-L "The Liverpool List" is run by the Department of Philosophy, University of Liverpool https://www.liverpool.ac.uk/philosophy/philos-l/ Messages to the list are archived at http://listserv.liv.ac.uk/archives/philos-l.html. Recent posts can also be read in a Facebook group: https://www.facebook.com/PhilosL/ Follow the list on Twitter @PhilosL. Follow the Department of Philosophy @LiverpoolPhilos To sign off the list send a blank message to philos-l-unsub...@liverpool.ac.uk.