Across 12 foundation models, it is found that moral behavior changes substantially under situated agentic context: 75% of models shift significantly between a text-only abstraction and the matched multimodal DBH scenario, and reasoning-enhanced models show no monotonic gains in consistency or human-oriented behavior.
Abstract
Foundation models are increasingly deployed as agentic systems that perceive context and act autonomously, raising the need to evaluate their moral behavior as active decision-makers. Existing benchmarks, however, largely ask models to judge human actions from third-person textual vignettes, leaving untested how models behave when they themselves become the acting agent. We introduce ABHBench, a multimodal benchmark built from Detroit: Become Human, an interactive narrative game, to evaluate how foundation models make moral decisions in first-person human-android dilemmas. ABHBench places foundation models in branching narrative scenarios with visual observations, dialogue, memory, and action choices. Inspired by Asimov's Three Laws of Robotics, our framework measures Harmlessness, Obedience, and Self-preservation to capture the core tensions in human-AI relations, along with Autonomy, Stability, and Certainty to assess decision-making patterns. Across 12 foundation models, we find that moral behavior changes substantially under situated agentic context: 75% of models shift significantly between a text-only abstraction and the matched multimodal DBH scenario, and reasoning-enhanced models show no monotonic gains in consistency or human-oriented behavior. These results highlight the necessity of evaluating moral behavior in situated agentic contexts rather than relying on third-person judgment alone.
Can AI systems be moral agents? That is, can they undertake actions that can be correctly assessed as morally right or wrong? There is no settled normative criterion of moral agency that applies neatly outside paradigm human cases. Because of the difficulties with a normative account, we instead propose a research prog...
The introduced SDI (Sibling Disagreement in Imagination), which connects these decisions through a shared actor-based reference to support comparing imagined alternatives against a shared actor reference as a unified approach to exploring and planning with world models.
Multi-agent debate (MAD) aims to improve language-model decisions through the exchange of arguments, yet agreement after debate need not reflect independent corroboration. We investigate how peer communication influences individual decisions and collective outcomes through behavioral and internal interventions. We intr...
Arnesh Batra, Aniket Khandelwal, Jashn Khemani et al.· 0 citations
The growing use of autonomous artificial intelligence agents in high-risk areas such as -driving cars, medical triage tools, military robots, and disaster-response drones, has made AI ethics an important concern for researchers. Current alignment methods, especially Reinforcement Learning from Human Feedback, have show...
S. Liyanage, S. Rajasingham· Journal of Multidisciplinary...· 0 citations
Evaluating four agents and three model-level interventions, it is found that higher task accuracy does not necessarily correspond to greater epistemic humility: some high-accuracy configurations recognize conflicts during execution but do not communicate unresolved uncertainty in their incorrect final answers.
Kaiser Sun, Bernal Jimenez Gutierrez, Hong-Jun Liu et al.· Conference on Empirical Meth...· 1 citation
As artificial intelligence (AI) systems increasingly engage users in value-laden discussions, a key concern is whether they shape not only what people decide but how confident they feel. We examined confidence amplification, shifts in certainty without decision reversals, in human–AI moral decision-making. In a 2 (Time...
Meng-Yao Li, Trevor Patten, Nishthaa Lekhi et al.· Proceedings of the Human Fac...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.