arXiv:2605.27622v1 Announce Type: new Abstract: To safely interact with humans, AI agents must both know our norms and consider them during planning. However, such norm-guided planning has been less explored, only within communities of artificial agents, and has ignored the dynamic nature of norms. This paper instead presents an approach to guiding planning with dynamically changing norms in a human-AI setting. We contribute a defeasible calculus for resolving normative conflicts and an approach to using such dynamically changing norms as guard rails on plans. We theoretically demonstrate our
The increasing deployment of AI agents in human-centric environments necessitates robust mechanisms for ethical and safe interaction, driving research into dynamic norm integration.
This research addresses a critical safety and trustworthiness challenge for AI, ensuring agents can adapt to complex human social structures and ethical considerations.
AI agents will be able to incorporate and resolve conflicts within dynamically evolving human norms, moving beyond static ethical frameworks.
- · AI developers
- · Robotics companies
- · Ethical AI researchers
- · Sectors deploying AI in complex human environments
- · Developers of rigid, non-adaptive AI systems
AI agents will exhibit more context-aware and socially acceptable behaviors.
Public trust and adoption of autonomous AI systems could increase significantly in professions and daily life.
The definition of 'AI ethics' may evolve from static rules to dynamic, adaptive frameworks, enabling more nuanced human-AI collaboration.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.AI