AI and Data Literacy: Challenges and Opportunities for Teachers and Students

This article explores the evolution, impact, and challenges of artificial intelligence (AI) in education. It defines AI and machine learning, tracing their historical development from the Turing Test to modern Large Language Models (LLMs) like ChatGPT and Gemini. The discussion highlights AI’s transformative role in education, emphasizing the need for data literacy and AI literacy among educators and students and introduces the AI Literacy Framework, outlining key competencies necessary for effective AI integration in learning environments. Additionally, the article critiques traditional educational methods in the AI era, advocating for experiential and personalized learning approaches. While AI presents both opportunities and risks—such as ethical concerns and academic integrity issues—the article underscores the irreplaceable value of human intelligence, creativity, and critical thinking.

Introduction: What is AI?

According to Russell and Norvig (2020) in their book Artificial Intelligence: A Modern Approach, AI requires an empirical science such as psychology, including observations and hypotheses about actual human behavior and thought processes. It also involves a combination of mathematics and engineering, while at the same time linking to statistics and control theory (Russell & Norvig, 2020). Attempting to define AI, it can be said that AI refers to the ability of a machine to reproduce cognitive functions of a human being, such as learning.

But what is learning? Learning is a process “by which an animal (human or non-human) interacts with its environment and becomes changed by this experience so that its subsequent behaviour is modified” (Hall, 2003). In the case where the subject is a computer, we call it machine learning: a computer observes some data, creates a model based on the data, and uses the model both as a hypothesis about the world and as software that can solve problems (Goodfellowet al., 2016).

AI: History and Evolution

Focusing on key moments in AI history, as shown in the Fig. 1 (Buchiokonicha, 2024), we note the “Turing Test” in the 1950s (Turing, 1950), the first appearance of the term AI in 1955 (McCarthyet al., 1955), and later developments in machine learning. Funding hiatuses in the 1970s and 1980s, known as “AI Winters” (Crevier, 1993), were followed by advances in deep learning and convolutional neural networks in the 1990s (LeCunet al., 2015). The proliferation of AI applications such as Apple Siri, Google Assistant, and Amazon Alexa further accelerated AI’s influence.

Fig. 1. History of artificial intelligence: How it all started (Buchiokonicha, 2024).

Large Language Models

Today, tools like ChatGPT (OpenAI, 2023), Gemini (Google DeepMind, 2024), and DeepSeek represent the next iteration of deep learning, referred to as Large Language Models (LLMs). These models generate information based on extensive datasets and refine themselves through machine learning cycles. However, they often lack verifiable sources, leading to issues of reliability (Benderet al., 2021).

Through these models with their great customization according to the needs of each scientific field, generating information from the huge amount of data they have access to in real time is an easy and productive process. This same process is repeated and constitutes the new training material of these models for their self-improvement, as machine learning besides, as it dictates, but with one assumption: The information produced often lacks source support and results in questionable reliability. Alongside the rapid development of tools that allow increasingly open access to data, education has to face new challenges and adapt to a new reality.

AI and the Educational Community

Teachers, students and their parents wonder what learning can mean anymore when software can answer every question. However, AI offers a new and exciting form of learning and an ideal opportunity to redefine the future of education.

It is a given through the history of technology that its democratisation brings about its development. From the time of Gutenberg’s invention of the printing press and the mass production of the book, from the time when computers were tools for people in white coats in a laboratory, and finally to the time when information can be accessed from home through the internet and more, through a smart phone, education has adapted and thus allowed each new generation of the respective era to become familiar with the technology that follows it and to exploit it to the fullest. So, it will be with AI.

But how can the educational community be equipped with the knowledge needed to adapt to this new reality? In previous years there was talk of Digital Literacy (Gilster, 1997) when computers were introduced into the educational process. But who was the digitally literate teacher? The one who was able to make use of a computer for the requirements of organizing his lesson with basic skills of using a text editor or even a presentation tool with the next stage being the adaptation of his lesson to the new digital reality with tools specific to the lesson. But what happens now? Previously we mentioned the large amount of data that AI has access to. The new concept that AI is introducing is Data Literacy.

As we distinguish from the Digital—Data Literacy pyramid in Fig. 2 (Bell, no date) the ability to read, analyze, transform, evaluate and apply data is what is called ”data literacy. Although data literacy for some can be difficult term to define, the idea of data literacy evolves as new technologies and data are made available. Data literacy is an underlying component of digital skill, someone’s ability and desire to use existing and emerging technology to drive better outcomes (Ghodoosiet al., 2023). Furtermore, helps bridge the gap between technical and non-technical team members. When everyone in an organization can understand and discuss data, communication and collaboration improve, leading to more effective teamwork. We can increase access and interest in data science careers by incorporating data skills into out-of-school learning programs. From the next figure, we can see that AI is closely related to data literacy: It allows a literate data user to be able to evaluate the reliability, validity and provenance of information as we can distinguish from the Fig. 3 (Correlation one, no date).

Fig. 2. How to improve data literacy as a leader (Bell, no date).

Fig. 3. How are data science, data literacy, and AI different? (Correlation one, no date).

To the question of how the educational community can be equipped with the necessary skills, the answer is the “AI Literacy Framework” or simply the AI skeleton as we see in Fig. 4 (Kennedy, 2023). Specifically, we distinguish seven elements:

Fig. 4. AI literacy framework (Kennedy, 2023).

1. Application knowledge.

2. Information and data literacy.

3. Communication and collaboration.

4. Content creation.

5. Security.

6. Problem solving.

7. Successful use and exploitation of AI tools.

In addition, the updated European Digital Competences Framework (DigComp) emphasises the need to develop skills to search, evaluate and manage digital data effectively. We note here that this includes an awareness that the data on which AI depends may contain erroneous data and that AI algorithms may be configured to provide the information desired by the user, and in our case, the teacher and learner, rather than what is actually valid.

Change is almost always viewed with mistrust and caution, or even outright negativity. One way to combat the change introduced by AI involves the use of software that detects text produced using AI. However, at the same time, other software is being developed which aims to deceive these detectors and give AI texts a convincing human face. Teachers are also required to monitor their students closely to prevent copying from AI, but this is time-consuming, if not impossible. Moreover, it is contrary to the teacher’s mission to build trust, accountability, independence and passion for learning in students.

The Scaffolded AI Literacy (SAIL) Framework for Education (MacCallumet al., 2024) defines six categories of the framework and is divided into three domains of AI literacy: AI concepts, application of AI and technical skills, and AI digital citizenship. Education must adapt to AI developments. The European Digital Competences Framework (DigComp, 2023) highlights the need for data literacy, which encompasses the ability to evaluate the reliability, validity, and provenance of information. We can define the “AI Literacy Framework” for educators identifies seven key elements: application knowledge, information and data literacy, communication and collaboration, content creation, security, problem-solving, and the effective use of AI tools (UNESCO, 2023).

And not only teachers but also institutions, as UNESCO states, publicly available Generative AI (GenAI) tools are emerging and being updated at a rapid pace that exceeds the pace of adaptation of any national regulations, the absence of which leaves educational institutions largely unprepared to validate these tools.

It is quite clear that this technology is popular, works and is cost-effective. That means it’s here to stay. We should not ignore issues relating to privacy, intellectual property rights, credibility, propaganda and the difficulty of managing these issues (Brownet al., 2020). However, what is undeniable is that AI is getting better with time.

If educators and parents are concerned with issues of academic integrity that could lead students into a context of bias and misconceptions and even bypass the learning process altogether, let us recall the years past when many feared that calculators and automatic spell check could destroy our students’ abilities to spell and do facts, but today such resources are taken for granted.

Traditional Teaching Methods and AI

AI challenges traditional education by questioning how learning is measured. Computers excel at tasks like essay writing and problem-solving but do not reflect the diversity of human intelligence (Wing, 2006). Teachers are encouraged to focus on experiential learning, active learning, personal relevance, real-world applications, emotional intelligence, and out-of-the-box thinking to maintain human-centric education (Kolb, 1984). AI in educational practice leads to the fundamental question of “how do we measure learning” since computers and devices can write essays, create diagrams and pictures, write code better and get A’s in exams. This means that the real threat to student education is not machine learning and AI. It is the way we have come to measure and evaluate human learning and human intelligence.

The problem with these traditional definitions of learning and intelligence is that they favor this kind of process since that is what computers excel at. Machines could beat humans in traditional learning tasks because for these kinds of tasks they can access and analyze more information much faster than humans. However, these definitions do not reflect the diversity of students’ thinking as they each think differently and contribute different perceptions and experiences to the classroom. When we assign tasks that aim to memorize and recall information, it is not surprising that students turn to AI for answers.

This criticism highlights a key question in the field of AI: Is it enough for a machine to imitate the behavior of a human to be considered intelligent, or does it need to have real inner understanding and consciousness? This is a complex question that depends on the definition of intelligence and consciousness and is causing considerable debate in the scientific community. We can argue that AI is obviously not the same as human intelligence.

Although AI can mimic or exceed human performance in specific tasks, there are substantial differences between the two:

Experiences and Emotions: Human intelligence is enriched by personal experiences and emotions that shape how we think and make decisions. AI, on the other hand, operates based on data and algorithms, without the ability to feel emotions.

Creativity and Inspiration: Humans possess the capacity for creativity and can generate brand new, original ideas out of thin air. Although AI can “create” based on programmed instructions and large amounts of data, this process does not compare to human creativity and inspiration.

General Intelligence and Agility: Human intelligence is capable of processing information from different knowledge domains, learning new skills, and adapting to change with flexibility. AI, especially in its current form, is specialized in specific tasks and lacks the general flexibility of human intelligence.

Critical Observation and Metacognition: Humans have the capacity for self-observation, critical examination and can make their own thoughts, sensations and learning processes (metaphor) subject to analysis. AI does not have this capacity for self-reflection or self-improvement in the same sense (Sofoset al., 2024).

From the above definitions we can suggest to teachers to redefine what intelligence is. The optimistic news is that the appropriate teaching methodology already exists in teachers’ toolboxes. That is to design teaching practice with the following five characteristics:

First, let’s leverage experiential learning and ask students how their actions influenced an outcome, or how the group worked together towards a common goal. Let’s motivate them to value problem solving through independent discovery, exploration, and reflection. This is part of what we call active learning, which we know is more effective than previous instructional models such as lectures, worksheets, and memorizing information from a book.

Second, let’s ask questions about meaning, let’s encourage students to clarify their beliefs. Let’s make learning personal, relevant and personalized by asking what the information means to the students or what it might mean to the community. And when students ask teachers why they need to learn something, there should be a convincing answer.

Third, let’s ask students to make connections to the real world to integrate the information they already possess. Studies show that students are more engaged when they understand the context and make connections to prior information, for example, how it relates to the previous week’s lesson or what they are learning.

Fourth, let’s encurage students to talk about their feelings, and cultivate empathy that puts them in someone else’s shoes, i.e., cultivating emotional intelligence. AI is still unable to draw information and conclusions about certain situations, such as human mood. In contrast, a one year old child, even a cat or a dog can read someone’s mood better than any software.

Fifth, let’s have students direct their knowledge and skills toward new applications. Apply current data to a new situation or propose a new way of looking at an old problem in a way that only humans can think, what we call “out of the box thinking.” AI can only follow the rules given to it and cannot think abstractly or understand the relevance between concepts that have no obvious connection. Thus she cannot access the perceptions, interpretations and feelings of students.

Conclusion

Teachers should utilize the “AI Literacy Framework” and develop AI-related skills while prioritizing human intelligence. AI should complement, not replace, human creativity and critical thinking (Brynjolfsson & McAfee, 2017). Therefore, teachers are encouraged to keep up with developments and incorporate new learning models into their teaching as much as possible. Most importantly, however, it remains important to value human intelligence more than any digital data processing and to push students to use aspects of human intelligence that machines do not possess. Asking questions that computers cannot answer which is done by challenging students with authentic experiences and tapping into the incredible capacity, diversity and creativity of the human mind.