Preparing Students for the Future: Teaching AI Literacy

As AI tools and models become increasingly available, students will likely be expected to make use of them in a range of academic and professional settings. AI will also likely become a core part of common tools, with the line between non-AI and AI features becoming increasingly blurred. Faculty can develop courses to help students build the digital literacy skills that will be required to engage technology of every kind. In-class and independent assignments can guide students to master (and think critically about) prompt engineering, as well as the quality of the content that AI tools generate.

Laupichler et al., (2023) describe AI literacy as: “the ability to understand, use, monitor, and critically reflect on AI applications without necessarily being able to develop AI models themselves.” This definition suggests that all students, in any discipline, would do well to understand:

• the historical context of generative AI;
• the current ethical debates surrounding generative AI;
• the current applications of generative AI in various disciplines and industries;
• and critical frameworks that help assess AI tools and applications.

Educators aiming for AI literacy should touch on each of these areas, though they may choose to dive into one or more of the areas with more detail. At minimum, students who possess AI literacy will understand that generative AI is not a new invention, be able to describe some main ethical (and legal) challenges surrounding the technology, identify a few of ways AI is being deployed in their field, and demonstrate the ability to use this knowledge to evaluate AI tools and applications.

Depending on the focus and scope of their  class, the educator could offer a summary of numbers 1-3, or they could dive into explicit lessons and discussions on each topic. Some educators might wish to refrain from having their students use generative AI as part of their coursework, but that does not mean the students cannot become fluent in the language and function of generative AI as we understand it today. They can touch upon or explore the long history of generative AI as it has progressed from ELIZA to Copilot, for example, and have students debate ethical usage in different settings. Students should leave their courses with, at minimum, a brief understanding of what generative AI is and how it is used in the discipline/field, regardless of how they practically apply it within the course.

The University of Florida has developed an “AI Across the Curriculum” initiative to provide all students the opportunity to develop their AI skills for future workforce participation. This will be accomplished through a combination of foundational courses, including “Fundamentals of AI” and “Ethics, Data, and Technology”. They’re also offering disciplinary courses such as:

• “AI in Agricultural and Life Sciences”
• “AI in Social Sciences”
• “AI in the Built Environment”
• “AI in Media and Society”

This means students have the ability to earn a certificate in AI to complement their degree. This approach does have the  downside of  requiring adding on several stand-alone courses to a student’s degree. However, consideration of the learning outcomes identified in the University of Florida’s program or other similar models might help institutions or programs find opportunities to infuse AI across the curriculum in existing courses

However, even without AI directly integrated into the curriculum, there are already some resources that are available to students. For example, the University of Sydney has created a great website called AI in Education, which is a guide for students (created by students!) with information like creating resumes, understanding content, and overcoming writer’s block. There will certainly be more information like this for students in the future, which is one of the reasons it’s so important to be transparent about these tools now.

Case Study: AI in Design

One example of a field that has already seen the impact of AI is the field of design, which is rapidly evolving. While generative AI tools such as Midjourney and DALL-E (AI that generates images from prompts) have only begun to grab the attention of the general public recently, AI design tools have been used in the industry for some time. For example,   Midjourney, Netflix’s AVA (aesthetic visual analysis) is a collection of tools and algorithms which encapsulate the key intersections of computer vision combined with the core principles of filmmaking and photo editing. This AI is being used to create the thumbnails and trailers based on users’ interactions with the content on the platform. Similarly, Alibaba’s AI tool, Luban, is capable of creating eight thousand customized banner ads in one second. Even before the hype AI is currently receiving, many everyday interactions were created by a small design team using AI tools (or, in many cases, created entirely by AI).

In addition, AI is being used to automate many tasks that were traditionally done by designers, such as generating layouts, creating color schemes, and even coming up with new ideas. Here are three examples of different AI tools in design and what they can do:

• Adobe Photoshop’s Content-Aware Fill uses AI technology to select and blend the best replacement pixels.
• Adobe Firefly uses generative AI to create many kinds of free art.
• Figma’s AI Designer plugin can streamline and enhance the user experience design process.
• Topaz Gigapixel AI can upscale and enhance image detail and resolution by 600%.

Now designers have more flexibility to tackle even more important tasks beyond design production. AI will not replace designers, but it will change the role of designers in the future. Designers will need to bring creativity, critical thinking, and problem-solving skills to the creative process, but they will also need to be proficient with AI tools and technologies.

AI Literacy in Everyday Living

An important component of AI literacy is awareness of the multiple ways AI is impacting the lives of students today — both within and beyond their college education. This is important to us as well, and not only as educators. AI is quickly becoming pervasive.

There have already been significant AI contributions in such disparate fields as arts and entertainment, climate change, cybersecurity, disabilities, healthcare, infrastructure, jobs, mental health, politics, and more. Examples can be a powerful aid to understanding. These areas are likely to continue to be foci of AI growth, though the specifics will grow over time:

Climate Change

With its abilities to classify and predict, AI is used in many ways that involve climate change. It has been used to predict future trends, forecast global temperature changes, and anticipate phenomena such as El Niño (Cowls et al.). It should be noted there are also concerns regarding the carbon impact of AI development and usage on our environment (Keller et al., 2024).

Smartphone use

AI is making a difference in our use of smartphones in many ways. In addition to voice assistants (e.g., Siri, Google Assistant), it improves photography by responding to the environment (such as low light)  provides real-time language translations, and more (Qualcomm Technologies, Inc., n.d.).

AI Uses in Smartphones

Mental Health

AI has made contributions to several aspects of mental health. Chatbots have been found to play a role, particularly in areas such as depression and anxiety. Woebot, developed at Stanford University, is such a bot with responses limited to recognized treatment approaches such as Cognitive Behavioral Therapy (Woebot Health, 2024). Findings support the suggestion that patients can form bonds with these conversational agents and show improvement (Darcy A et al., 2021).

They may even have particular benefits for college students (Maples et al., 2024) and the elderly (Valtolina, Stefano & Hu, Liliana, 2021). Bots can help to provide companionship, reminders, and general support.

Other studies looked at the diagnosis of psychiatric disorders. AI can be useful in their  prediction, diagnosis, treatment, and monitoring (Terra et al., 2023). For example,  AI’s analysis techniques contribute to the value of MRI results as well as of EEG signals used in the diagnosis of psychiatric disorders (guang-Di et al., n.d.). With its growing abilities, AI can be, and is likely to become increasingly, useful in the recognition of emotions, improvement of communication during treatment,  identification of high risk of suicide, and prediction of remission after treatment, among others (Terra et al., 2023).

Shopping

Our online shopping experience is now shaped by AI in several ways. It helps personalize our shopping experience and make it more efficient. For example, an Amazon order can now go from being placed to readied for delivery in 11 minutes (Van Cleave, & Novak, 2023).

Healthcare

AI’s contributions go well beyond mental health. It has quickly become a vital part of our healthcare system (Pandey, 2024). From early detection through AI’s image recognition skills, to improved diagnostics through its ability to quickly analyze massive amounts of data, to precision in robotic surgery, the improvements in healthcare have been felt throughout the field.

Entertainment

Streaming services like Netflix provide recommendations about what users would like to view next by analyzing behavior and previous choices. AI is also involved in other ways, for example optimizing streaming by determining how best to encode the video so as to maximize quality while minimizing file size (Kjandelwal et al., 2033).

Everything mentioned here is likely to become obsolete soon, replaced by even more powerful contributions to our lives. As we have discovered, unfortunately, there’s another side to it.

Misinformation and Disinformation

Clearly there are benefits to artificial intelligence, but its pitfalls can be deep. Deepfakes are now part of our culture and are becoming increasingly realistic. We’ve already seen them in several areas ranging from misinformation in entertainment such as the Tom Cruise deepfake you may have seen (Inspired Learning, 2022) and the Jordan Peele video of Barack Obama to disinformation in politics such as  the Joe Biden robocall (Bushard, 2024; Satariano & Mozur, 2023; Vincent, 2018).

How will we deal with this?  Will technology ultimately provide a solution to the problem it has created?  A difficult question confronts us as educators (and human beings) — how do we learn, and teach, when to not trust our eyes? It may be that technology will offer solutions, but in an area with such rapid development, it’s unlikely those  solutions will remain effective over time. How do we unlearn a habit we’ve been using throughout our lives? That social experiment is underway.