TY - JOUR
T1 - Empowering learners for the age of artificial intelligence
AU - Gašević, Dragan
AU - Siemens, George
AU - Sadiq, Shazia
N1 - Funding Information:
This special issue was in part supported by funding from Australian Research Council ( DP210100060 , DP220101209 ), Economic and Social Research Council of the United Kingdom ( ES/S015701/1 ), and Jacobs Foundation (CELLA 2 CERES) awarded to the first author. Any opinions, findings, conclusions, or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the Australian Research Council, Economic and Social Research Council of the United Kingdom, or Jacobs Foundation.
Funding Information:
The first and most prominent theme centres on what we describe as “the space of coordination” - the intersection between AI and humans. What is being coordinated in this space? Currently, almost everything related to learning is under examination. Molenar (2022) focuses on tasks that can be “offloaded from humans to AI” (p. 1) and then “onloaded from AI to humans” (p.1), focusing specifically on a hybrid systems to help develop self regulated learning skills in young learners. She offers a hybrid model of regulation, emphasizing four regulation degrees: AI, co-regulation, shared-regulation, and self-regulation with an intent to support SRL developing through interaction with AI. Markauskaite et al. (2022) explore this middle space by engaging a number of experts in the learning analytics and AI fields to articulate some direction of an “interconnected, fast changing world” where “agentic machines” (p. 1) are the norm. Their Table 1 captures the range of perspectives regarding what capabilities are needed, how we develop those capabilities in humans, and how we study the development of those capabilities. Siemens et al. (2022) suggest that intelligence is too broad a term to provide utility when assessing which specific tasks or activities should be conducted by human or machine. Instead, they argue that cognition and discrete cognitive activities should be the foundation for evaluating and assessing how people learn, sensemake, and make decisions when agents with cognitive capacity are present.The third major theme of this issue relates to explainability of AI. Explainability refers to the ability of humans to understand and trust AI. This need will increase in importance, especially as the intersections between human and AI require greater collaboration and coordination (see Carvalho et al., 2022; Markauskaite et al., 2022; Molenar, 2022; Siemens et al., 2022). Khosravi et al. (2022) introduce the XAI-ED, detailing how existing explainable AI approaches can be applied to the education sector. XAI-ED incorporates insight from numerous fields, including cognitive and learning sciences, AI, human computer interaction, and learning analytics in order to present a “means for educational tool developers and researchers” (p. 4) with the means to support the “creation of trustworthy, AI-augmented, sociotechnical systems” (p. 18). They then present four case studies that detail how XAI-ED is reflected in different AIED systems: learner sourced, open-ended learning environments, writing analytics, and team-based learning. They conclude by advocating for broader discussion around explainable AI in education, increased research and adoption of these models, and involvement of multiple stakeholders in developing tools and systems.This special issue was in part supported by funding from Australian Research Council (DP210100060, DP220101209), Economic and Social Research Council of the United Kingdom (ES/S015701/1), and Jacobs Foundation (CELLA 2 CERES) awarded to the first author. Any opinions, findings, conclusions, or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the Australian Research Council, Economic and Social Research Council of the United Kingdom, or Jacobs Foundation.
Publisher Copyright:
© 2023
PY - 2023
Y1 - 2023
N2 - Although studied for decades by the research community, artificial intelligence (AI) in education has recently sparked much public debate with the wide-spread popularity of systems such as ChatGPT and DALL-E. Existing literature offers a wealth of research on design, deploying and evaluating AI-driven systems in education. However, the challenges related to the growing influence of AI in the society, calls for revisiting research foundations of AI in education in order to inform decision-making in policy and guide future research. This special issue of Computers & Education: Artificial Intelligence brings 11 papers that explore the theme of empowering learners for the age of AI. The contributions of these papers can be broadly grouped into seven main themes: intersection between AI and humans that looks at the space of coordination; assessment that explores challenges and opportunities afforded by the use of AI in educational assessment; explainability in AI as a critical need for humans in education to understand and trust AI; design for learning that offers principles for designing AI-driven systems and educational opportunities; conceptual AI and learning exploring the need for the development of new theories of learning and their connections with existing theoretical foundations in education; accurate predictions and their role in future education; and applications of AI in classrooms and educational systems. The findings of these studies highlight pressing research and policies challenges and opportunities that arise with the broad penetration of AI in education. They also emphasize the need for future research that addresses issues of ethics, bias and farness in the use of AI in education; challenges associated with data sources and ownership as the key fuel and enabler of present-day AI generation; AI literacies and competencies of stakeholders who use and are impacted by AI in education; identification of effective learning and teaching practices with the use of AI; and policy development to increase responsiveness of education systems to rapid changes driven by AI.
AB - Although studied for decades by the research community, artificial intelligence (AI) in education has recently sparked much public debate with the wide-spread popularity of systems such as ChatGPT and DALL-E. Existing literature offers a wealth of research on design, deploying and evaluating AI-driven systems in education. However, the challenges related to the growing influence of AI in the society, calls for revisiting research foundations of AI in education in order to inform decision-making in policy and guide future research. This special issue of Computers & Education: Artificial Intelligence brings 11 papers that explore the theme of empowering learners for the age of AI. The contributions of these papers can be broadly grouped into seven main themes: intersection between AI and humans that looks at the space of coordination; assessment that explores challenges and opportunities afforded by the use of AI in educational assessment; explainability in AI as a critical need for humans in education to understand and trust AI; design for learning that offers principles for designing AI-driven systems and educational opportunities; conceptual AI and learning exploring the need for the development of new theories of learning and their connections with existing theoretical foundations in education; accurate predictions and their role in future education; and applications of AI in classrooms and educational systems. The findings of these studies highlight pressing research and policies challenges and opportunities that arise with the broad penetration of AI in education. They also emphasize the need for future research that addresses issues of ethics, bias and farness in the use of AI in education; challenges associated with data sources and ownership as the key fuel and enabler of present-day AI generation; AI literacies and competencies of stakeholders who use and are impacted by AI in education; identification of effective learning and teaching practices with the use of AI; and policy development to increase responsiveness of education systems to rapid changes driven by AI.
UR - https://www.scopus.com/pages/publications/85148750390
U2 - 10.1016/j.caeai.2023.100130
DO - 10.1016/j.caeai.2023.100130
M3 - Editorial
AN - SCOPUS:85148750390
SN - 2666-920X
VL - 4
JO - Computers and Education: Artificial Intelligence
JF - Computers and Education: Artificial Intelligence
M1 - 100130
ER -