University portals are essential digital platforms that support student engagement, resource access, and administrative processes. Despite their central role, achieving inclusivity in these portals remains challenging, particularly for international students and individuals with specific linguistic or accessibility needs. Research highlights persistent inclusivity gaps in current portal designs, which often lack adaptive features for diverse cultural, linguistic, and accessibility requirements (Mahrishi, Abbas, & Siddiqui, 2024; Rezwana & Maher, 2023). This study addresses these issues by proposing an AI-driven framework to enhance inclusivity in university portals, with Drexel University serving as a case study.

The framework leverages AI concepts, including techniques like pattern recognition and language adaptation, to inform adaptive features that support inclusivity. Natural Language Processing (NLP) is used conceptually for real-time translation to reduce language barriers, while machine learning (ML) is applied to detect behavioral patterns, enabling personalized adjustments that accommodate varied user needs without requiring direct technical implementation (Xu et al., 2024; Li et al., 2024). The application of AI in this context is also guided by ethical considerations, such as data anonymization and secure handling practices, to protect user privacy while maintaining transparency and trust (Winner, 1980; Lambert, 2018).

The methodology follows the Design Thinking approach, structured across five phases: empathy, definition, ideation, prototyping, and testing. During the empathy phase, user insights are gathered through surveys and interviews, revealing inclusivity challenges specific to linguistic and cultural adaptability. The ideation and prototyping phases generate and test adaptive features using Figma, demonstrating AI's potential to enhance usability and engagement across diverse user backgrounds (Becker & Yannotta, 2013; Gould & Lewis, 1985). Usability testing with a diverse group of users then assesses the framework's effectiveness in real-world scenarios, ensuring iterative improvements (Nielsen & Molich, 1990).

This research contributes to inclusive design by showcasing how AI-driven adaptive language support and personalized layouts can foster engagement and reduce barriers for non-native speakers and students with disabilities. The proposed framework aims to be scalable, offering universities a model for enhancing digital inclusivity that supports diverse student populations and aligns with ethical AI practices. Ultimately, this framework envisions a digital environment where all students have equitable access to educational resources, regardless of cultural or linguistic background (Shalamova, 2019; Quesenbery & Szuc, 2011).

Keywords: Inclusivity, AI, university portals, cultural adaptability, linguistic adaptability, accessibility, Design Thinking, ethical AI, educational technology.