This study investigates English as a Foreign Language (EFL) teachers’ perceptions of ethical artificial intelligence (AI) use in teaching, learning, and assessment within an Omani higher education context. Despite rapid AI adoption in education, institutional governance frameworks remain critically underdeveloped, particularly in EFL contexts — creating an urgent need for empirical, locally grounded research. Using a convergent mixed-methods design, quantitative data were collected from 52 EFL faculty members through a structured 52-item Likert-scale questionnaire, complemented by focus group discussions with nine purposively selected teachers drawn from the same participant pool. Analysis across five constructs revealed high levels of AI literacy and ethical awareness (M = 4.09), ethical responsibility and academic integrity (M = 4.19), and positive pedagogical engagement (M = 4.11). The most critical finding was a significant institutional policy deficit reflected in the lowest construct mean (M = 3.25), with the majority of participants reporting an absence of clear guidelines, consequences, or detection tools. Future orientation and framework acceptance recorded the highest mean (M = 4.30), with 98.1% of participants endorsing formal adoption of an Ethical AI Responsibility (E.A.R.) framework. Qualitative findings corroborated a persistent awareness–practice gap, student over-reliance on AI, and inadequate institutional scaffolding. The study recommends urgent development of context-specific, human-centered AI governance frameworks that bridge individual ethical awareness and institutional policy action, with particular relevance to Omani and comparable EFL higher education contexts.
GAOKAO-Bench is introduced, an intuitive benchmark that employs questions from the Chinese GAOKAO examination as test samples, including both subjective and objective questions that contribute a robust evaluation benchmark for future large language models and offers valuable insights into the advantages and limitations of such models.
Xiaotian Zhang, Chun-yan Li, Yi Zong et al.· arXiv.org· 216 citations· ⚡17
This work investigates the possibilities of using LLMs in a resume screening setting via a document retrieval framework that simulates job candidate selection and finds that the MTEs are biased, significantly favoring White-associated names in 85% of cases and female-associated names in only 11.1% of cases.
Empirically, PRISM reduces the end-to-end time for data selection and model tuning to just 30% of conventional pipelines, and achieves this efficiency while simultaneously enhancing performance, surpassing models fine-tuned on the full dataset across eight multimodal and three language understanding benchmarks.
Jinhe Bi, Yifan Wang, Danqi Yan et al.· arXiv.org· 73 citations· ⚡4
The method, ECCOLA, is presented, which aims at making the high-level AI ethics principles more practical, making it possible for developers to more easily implement them in practice.
Ville Vakkuri, Kai-Kristian Kemell, P. Abrahamsson· EUROMICRO Conference on Soft...· 64 citations· ⚡6
This paper designs Markov decision processes (MDPs) for different combinatorial problems and proposes to train conditional GFlowNets to sample from the solution space and demonstrates that GFlowNet policies can efficiently find high-quality solutions.
Dinghuai Zhang, H. Dai, Esmeralda S. Whitammer et al.· Advances in Neural Informati...· 59 citations· ⚡8
An empirical study on the current state of practice in artificial intelligence ethics is conducted by means of a multiple case study of five case companies, which indicates a gap between research and practice in the area.
Ville Vakkuri, Kai-Kristian Kemell, Joni Kultanen et al.· arXiv.org· 56 citations· ⚡6