Opening the Dialogue: How AI is Shaping the Future of Education

Will AI serve as a catalyst for innovation, or will it widen the gap between effective learning and mere efficiency? AI’s impact on education is profound and multifaceted, leaving us to ponder whether it will be our guiding light into the future or a dark tunnel with unforeseen consequences.

Because of this, universities need to respond not only by integrating AI into their policies and curricula, but also by thinking about how knowledge is delivered. This brings us to another question: of whether the way we teach is right for the 21st century. The discussion about teaching methods has been going on for a while, even before AI in the Czech context. One of the arguments related to instruction is that students are not really learning and gaining knowledge, but rather learning to decode or guess the “teacher’s password” and say the right answer to pass an exam and get credit without proper understanding.

Prof. Mazur has said in one of his speeches that lectures are problematic for many reasons, but the main flaw is that the traditional “lecture method is a process whereby the lecturer’s notes are transferred to the students’ notebooks without passing through the brains of either.” These shortcomings are exacerbated by the democratization in the use of AI because students might rely on AI-generated summaries rather than engaging with the material directly. However, with proper and sensitive use, AI could bridge these weaknesses and make the learning process more effective and attractive for both students and instructors.

This post seeks not to provide definitive answers, but is intended to spark a conversation about AI’s role in education.

Mastering AI: Literacy and the Art of Prompting

AI literacy refers to a set of skills that allow individuals to effectively and responsibly use artificial intelligence tools (Long & Magerko, 2020). In an era where AI is becoming increasingly embedded in educational processes, understanding these tools goes beyond mere usage. It involves knowing how AI systems make decisions, recognizing their limitations, and being aware of the ethical implications of their use.

The problem we face today is that many people using AI tools are not fully knowledgeable about how to use them effectively and responsibly. In other words, they are not aware of the limitations, capabilities, and processes that go on under the hood of these tools. Inadequate AI literacy among students and instructors can lead to ethical problems related to unintentional plagiarism, low-quality outputs, and even intentional academic misconduct.

One of the most critical aspects of AI literacy is mastering the art of creating prompts. This involves designing inputs or “prompts” that guide AI systems to produce useful and accurate outputs. In an educational setting, creating appropriate prompts can determine whether an AI tool enhances learning or simply generates generic and unhelpful responses. In other words, this is very well captured by the concept garbage in-garbage out. Working properly with AI should not be based on trial and error but rather on thoughtful steps and a clear idea of the desired outcome.

Mastering prompt creation and AI literacy is not without its challenges. It requires a clear understanding of both one’s subject matter and how AI systems process information. This is where universities should take a proactive approach. The trend is to actively promote the use of AI by students and instructors, but few, if any, universities have yet incorporated teaching AI literacy into their curricula.Related to this is the dilemma for universities of who will train the trainers (instructors) and how to deal with the generational skills gap, where students are sometimes several steps ahead of their instructors in terms of using AI tools.

AI in Education: Guiding Light or Dark Tunnel?

AI holds immense potential for enhancing the educational experience, offering personalized learning capabilities and increased efficiency in various tasks. It also promotes greater accessibility through tools that accommodate diverse learning needs, such as language translation, adaptive learning for varied paces, and assistive technologies for students with disabilities. However, this potential comes with risks—AI use could exacerbate inequalities, lead to over-reliance on the tools, and diminish the depth of human learning if not carefully managed.

When leveraged thoughtfully, AI can be a powerful tool for good. It can provide tailored educational experiences, broaden equitable access to knowledge and tools, and potentially help educators focus on more creative and meaningful aspects of teaching. By enhancing personalization and efficiency, AI can support students in reaching their full potential. For instance, integrating AI assistants into research databases like Scopus or Web of Science allows students and educators to navigate complex research landscapes more effectively, identify relevant studies, and streamline literature reviews, making high-quality research more accessible to a broader audience. Additionally, using AI tools has the potential to help reduce the inequality and discrimination faced by non-native English-speaking students and instructors in an English-dominated scholarly publishing environment. Banning the use of AI tools may do more harm than good (in fact, this would likely be futile) and a more realistic course should involve the creation of clear institutional strategies and guidelines for the use of AI, both in and out of academia.

There is admittedly a dark side to AI. If over-relied upon, it can, among other things, contribute to a lack of critical thinking, foster dependency, and perpetuate biases inherent in the algorithms. Without careful oversight, AI could lead to a future where education becomes more about surface learning than about truly understanding and engaging with educational material. Moreover, AI is entering a battlefield for information, media, and digital literacies for our students that has not yet been won yet.

The future of AI in education depends on the choices we make today. Educators and institutions must take an active role in shaping how AI is integrated into learning environments. This includes developing clear ethical guidelines, promoting AI literacy (and other literacies), and ensuring that AI tools are used to complement, rather than replace, traditional educational methods. Institutions must also determine who is responsible for specific steps in the use of AI tools for use in the academic context.

As we stand at this crossroads, it’s up to us to decide whether AI will be our guiding light or the beginning of a dark tunnel in education. By approaching AI with caution, curiosity, and a commitment to ethical practices, we can ensure that it enhances, rather than diminishes, the educational experience.

Will AI lead us to a brighter educational future, or will it cast shadows that are difficult to escape? The answer lies in our hands.

Resources:

Long, D., & Magerko, B. (2020). What is AI literacy? Competencies and design considerations. Proceedings of the 2020 CHI Conference on Human Factors in Computing Systems, 1–16. https://doi.org/10.1145/3313831.3376727

How do we better-equip ourselves and our students or mentees to deal with adversity and uncertain environments? No one has the right answer yet, it seems, but I came across two recent short essays which helped me start sharpening the way I’m asking myself these questions and thinking towards the future. 

Flateby, T. L. & Rose, T.A. (2021). From College to Career Success: How Educators and Employers Talk about Skills. https://www.aacu.org/blog/college-career-success-how-educators-and-employers-talk-about-skills

Even as we make curricula to non-discipline-specific learning outcomes for graduates such as the Association for American College and University’s VALUE (Valid Assessment of Learning in Undergraduate Education) rubrics, are we sure the outcomes we in academia create align to the expectations of the places graduates will work? Flateby and Rose (2021) describe early findings from a broader College to Career project, a small survey of line managers, which highlight gaps in graduates’ ability to think critically and communicate effectively:

Several managers observed that graduates need more experience exercising critical thinking skills throughout the curriculum and in more complex situations. Newer graduates often look for the “right” answer, the managers said, and provide employers with what they think they want to hear. Often, new graduates do not know how to proceed without direction.

For written communication, the line managers reported that newer graduates typically communicate in writing as though they are texting. Most of the written communication issues they identified pertained to a lack of audience awareness.

In response, Flateby and Rose suggest several immediate responses educators might take to address these issues in their assignments. 

But while assessing audience in writing may be easier to incorporate into assignments, “critical thinking” remains, despite an articulated VALUE rubric, a rather nebulous concept. 

Burke, T. (2021). An Unconvincing Argument for the Liberal Arts: We say we prepare students for undefined futures. Are they better for it? https://www.chronicle.com/article/an-unconvincing-argument-for-the-liberal-arts (NOTE: while this essay is freely-available, it is behind a “data wall” and one must provide one’s contact information to view it). 

Burke discusses how hard it is to put our educational fingers on what we are guiding our students to think critically about, particularly in uncertain times:

Our assumptions about how to teach to uncertainty are mostly unexplored, and the empirical evidence of whether we do so successfully is debatable. To the degree that we are successful, we don’t really know why. Arguably, the capacity to navigate uncertainty has less to do with student learning than with the social capital and economic resources available to our graduates. This is where “preparation for uncertainty” lives alongside other reassuring concepts like “resilience,” “emotional intelligence,” or “grit.” These concepts may not be measuring teachable skills or habits of mind so much as access to money and social networks. Dealing with rapidly changing conditions is much easier if your parents can help with the rent or if you know someone who can get you in the door in a new line of work after your current gig closes down.

Many of us would answer “critical thinking” (which may be an equally leaky terminological boat). We’d likely assert that critical thinking suffuses our institutions in such a way that their graduates learn to view the world around them skeptically and provisionally, and that this in turn prepares them to adapt rapidly to changing economic and social conditions (and to help lead or direct processes of change for others). The major problem with this answer is that any curricular structure, any pedagogy, can likely and perhaps justifiably claim to be producing critical thinking and hence to be preparation for uncertainty. It’s so truistic and underspecified that it’s hard to be satisfied with it as an answer. Possibly, we could decompose “critical thinking” to far more specific epistemological and methodological commitments in various academic disciplines: the scientific method, thought experiments, close reading, etc., and get a better account of how to teach skepticism, provisional truth-making, and so on. Possibly.

Read Burke’s concluding thoughts about this topic, and read them again, particularly the entire last paragraph of his essay. His contemplation don’t provide a clear pathway towards the future, but they do crystalize some thoughts many of us in the educational sector have been considering these past few months:

Inhabiting the foundational uncertainty of the universe is one of the deepest challenges of human life. If we have insight into that, good. If we don’t, let’s work to develop that insight. But we mustn’t confuse this work with the drive to normalize the insecurity of our present moment. Our educational job there is different: We must teach our students to reject that project entirely.

[W]e must be especially sensitive to the ways in which open science policies may actually worsen existing inequalities, and make efforts to mitigate these effects.

Read more:

Ross-Hellauer, T., Fessl, A., & Klebel, T. (29 December 2020). Can We Have Open Science Where No Scholar Is Left Behind? Social Science Space. https://www.socialsciencespace.com/2020/12/can-we-have-open-science-where-no-scholar-is-left-behind/

Besides simply the number of classes represented here, I think another point this data makes well is how spread across disciplines these classes are. About 50 are taught in computer science departments, but the rest come from information science/studies, communication, law, philosophy, and others.

Fiesler, C. (2018). Tech Ethics Curricula: A Collection of Syllabi. Retrieved from: https://medium.com/@cfiesler/tech-ethics-curricula-a-collection-of-syllabi-3eedfb76be18