AI-ALOE RENEWAL: AI for Skilling, Reskilling, and Upskilling Adult Learners: From Online Education to Workplace Training

Budget: $ 18.4M. Award period: 01/01/2027 – 12/31/2031 The AI Institute for Adult Learning and Online Education (AI-ALOE) addresses the growing disruption of worker skills by leveraging cutting-edge, responsible artificial intelligence to support the continuous skilling, reskilling, and upskilling of adult learners throughout their careers. Grounded in theories of human cognition, learning engineering, and educator collaboration, AI-ALOE’s next five-year initiative focuses on building Generative and Agentic AI technologies designed to foster adaptive expertise in workplace and higher education domains such as advanced manufacturing and data science. Backed by an interdisciplinary team of over 20 scientists, robust industry relationships, and major educational partnerships, the institute aims to reach at least 250,000 adult learners, train over 100 Educational Innovation Fellows, and serve as a national nexus for AI-augmented workforce development.

AIM HIGH: AI-Powered Multimodal Hybrid Instruction for Growth

AIM HIGH (AI-Powered Multimodal Hybrid Instruction for Growth) is a scalable, multi-agent platform designed to advance hybrid learning through AI agent-supported, one-on-one conversational coaching. The platform hosts a suite of learning agents that support knowledge exploration, knowledge construction, and knowledge demonstration—capabilities essential for upskilling and reskilling workforce competencies. Currently, AIM HIGH includes agents designed for summarization, causal/process analysis, and argumentation activities, with its companion system.

AI-Powered Multimodal Analytics of Learner Dynamics in Learner-Centered Activities

Learner-centered learning environments encourage autonomy, collaboration, and creative problem-solving, but they also present challenges in understanding how students engage and progress individually. This project leverages AI-driven multimodal analytics to examine learners’ dynamics using video data collected from a learner-centered classroom, focusing on how learners interact with peers, tools, and tasks. By using computer vision, machine learning techniques, and multimodal large language models (MLLMs), the project aims to detect key events such as collaboration, focus shifts, tool use, and moments of guidance-seeking. The ultimate goal is to develop an AI model that supports timely and personalized feedback, offering instructors and learners deeper insights into the learning process and helping to shape more responsive educational practices.