Defeasible Conditional Obligation in a Two-tiered Preference-based Semantics (Extended Version)
arXiv:2604.26977v3 Announce Type: replace-cross Abstract: In response to a concern raised by Horty, this paper develops a two-tiered, preference-based semantic framework for modeling defeasible conditional obligations. The paper extends a Hansson-Lewis style preference semantics for dyadic deontic logic by incorporating a nonmonotonic reasoning mechanism that enables previously derived obligations to be withdrawn when new, potentially conflicting information comes in. The account is bi-preferential: two orderings--ideality and normality--on worlds are employed to address shortcomings in earlie
This paper addresses a known concern in deontic logic, reflecting ongoing academic efforts to refine the theoretical underpinnings of AI systems dealing with ethical reasoning and obligations.
For a sophisticated reader, this represents progress in formalizing ethical reasoning within AI, which is crucial for developing autonomous systems that can navigate complex real-world scenarios responsibly.
The proposed two-tiered, preference-based semantic framework offers a more robust method for AI to handle defeasible conditional obligations and conflicting information, potentially leading to more adaptable ethical AI.
- · AI ethicists
- · Developers of autonomous AI systems
- · Academic researchers in AI and logic
- · Developers relying on simpler, less robust ethical AI models
Improved theoretical models for AI ethical decision-making are developed and validated.
This foundational work enables the creation of more sophisticated and trustworthy AI agents capable of nuanced moral reasoning.
The broader adoption of such frameworks could lead to AI systems that are more dependable in legally and ethically sensitive applications, fostering greater public acceptance.
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