Dr. Matt Rhoads is a Tech and Instructional Leader and Innovator with hands in Adult Ed, K-12, and Higher Education. He is the author of several books and is the host of Navigating Education – The Podcast.
By: Matt Rhoads, Ed.D.
What is Artificial Intelligence (AI) Literacy? It is comprised of five major components that require us to teach students a wide variety of skills that are integrated with the very technology that has AI embedded in it. AI literacy and its themes encompass perception, representation and reason, learning, natural interaction, and societal impact. All of these factors relate to how computers collect data, interact with that data, and how we as humans can interact with it. These components of AI Literacy are described below as we discuss how AI is impacting our work as educators. From understanding what AI entails to how it impacts our workflow and our practice as educators, it is and will change how we operate.
Via AIK12s Five Big Ideas in AI
Note: AI Literacy still is being further defined by the greater community as it is an emerging technology that is now being scaled.
The goal of this article is to describe how AI works, how it is used in industry, in our own personal workflows, and in education. Additionally, the goal is to provide foundational information in each of these areas. Last, as you progress through this articles, think about the five themes discussed above in relation to AI Literacy. Think about how this may look in practical examples in your daily life as well as your students. Additionally, think about how you can teach about AI and how to use it in your content area and industry.
How AI Works?
AI encompasses a number of tools and mechanisms that analyze data to solve problems through algorithms. Tools and mechanisms such as machine learning, deep learning, neural networks, computer vision, and natural language processing can be utilized as a pathway to AI. For example, machine learning is a pathway that uses algorithms to learn insights and recognize patterns in data automatically. Then, it applies that learning to make better decisions over time based on the number of data processed and data available. Another example is deep learning, which is a more advanced method than machine learning that acts as large neural networks that function like the human brain to analyze data in a logical manner to find and learn complex patterns. Ultimately, its goal is to make logical predictions without the need for human input.
To see how AI works in your daily life, an example you see every day is text prediction. Using data inputted from users based on the words and phrases utilized over time, while writing a sentence the AI makes an informed prediction of what text may follow thereafter. Another example of this in action is predicting favorite websites, netflix show preferences, and advertisements we see. Based on the data we’ve provided over time along with users with similar preferences, through the nature of the algorithm, it can predict our preferences and even our behavior. For example, you will see your top preferences immediately in front of you to select. More often than not, someone will select that top preference reinforcing the prediction made.
Have you ever noticed how one preference or advertisement may be larger than the other? Developers can use what we call A/B testing to see user patterns over time, which amasses large data sets to see which preference users are more likely to press. This then reinforces the predictions the algorithms make and you will see how those same preferences are visible throughout your interactions on a variety of platforms (i.e., YouTube, Instagram, Netflix, Google Search).
AI in Industry
AI now has a presence in every industry. For many industries, the goal of utilizing AI is to improve efficiency, profitability, and productivity. For example, in banking AI can use machine learning algorithms to prevent fraud and cybersecurity attacks. Additionally, it can use biometrics and computer vision AI-based algorithms to authenticate users and their identity. For example, think about when you last logged into your online bank. Usually, you must first provide a biometric identification such as your fingerprint or face. Then, through what we call dual authentication, we are sent a text message to our phones with a randomized password, which then allows us to log in. Another example is in Healthcare, which can use our biometric data and historical biometric data to make predictions about our current and future health by taking vast amounts of data sets that appear to have similarities to ours. Through these predictions, doctors can see a number of prognostics they can then work towards with the treatment or preventative care.
Understanding AI Bias
AI has bias’ and is not entirely accurate. It is only as accurate as the data it utilizes to make ongoing predictions. Also, note that the data it may have in its database may also be biased. For example, when we use Google Search, our own searches have our own human biases, which Google then stores in its database. While algorithms may have rules built into them that try to filter biases within the data, it is imperfect as the biases may manifest themselves as the data it is making these predictions are inherently biased. Ultimately, this same principle also applies to the accuracy of the data it is pulling for its algorithms. Misinformation can easily be placed within a database that can be pulled by AI. Thus, bias and misinformation can be easily pulled into AI, which can then be scaled to meet mass audiences (i.e., social media newsfeeds).
With this said, we need to discuss bias can be filtered, but as discussed, we must have filters and safeguards in place for when the data is collected and then when it is pulled by the algorithms powering AI. While in many cases AI can reduce our own human subjective interpretations, it can scale them quite easily due to our own subjectivity found in the data that we are feeding it. Thus, we must determine how to measure fairness in terms of the data we are collecting and utilizing as well as pre-process data to counterfactual data to filter sensitive attributes.
In the meantime, we must be judicious and skeptical of all information we see. We must triangulate our conclusions by synthesizing the information we process as the accuracy of the information we encounter each day is subject to many biases’ and can contain misinformation. Therefore, a key skill to build is digital and multimedia literacy, which is also based on having a higher degree of literacy to process language and understand it. When thinking about any education program, we must keep these skills central to what we are doing.
Integrating AI into Workflow
We now live in a world where we have a personal secretary embedded into many tools we use every day. From text prediction, the impressive language and code applications of ChatGPT, formulas in Excel and Sheets, and predictive content to edit in tools like Canva, Adobe, etc, to the search capabilities of search engines, this is only the tip of the iceberg. How this can impact your workflow is astounding. Below are a number of examples of how AI can help improve workflow.
- Create templates of content to be used for email, newsletters, communication, marketing, and more.
- Revise writing to make it more engaging and grammatically correct.
- Providing code templates in a variety of different languages such as Python, Java, HTML, etc.
- Search content and provide basic information on topics
- Provide resources
- Generate multiple drafts of content
AI in Education
In our classrooms, we are collecting vast amounts of data on our students when they engage with our EdTech tools. From basic quizzes, time on task, and content they have created, to the number of clicks and the locations of those clicks in relation to the problem they may be solved, a plethora of data is being collected. Currently, AI is embedded in various tools, which predict student performance based on their current performance. For example, on tools like iReady, MobyMax, ReadTheory, and even state tests, adaptive AI can review large amounts of data based on present and student performance and provide a student with a question/problem at their current ability level. It can even provide lessons and tutorials based on student performance levels and determine which areas students need to make progress in before learning various skills that may be more difficult.
The implications of this go far beyond what was described above. AI can be harnessed by teachers to make real-time data-driven decisions for individual students, groups of students, and entire classes of students. We can then make decisions about our student progress and utilize assistive AI such as ChatGPT to support us in making personalized learning opportunities for our students.
Additionally, AI can improve the workflow of teachers by providing opportunities for them to optimize their workflow and instruction in their classroom through the following means:
- Brainstorming Ideas.
- Revising Writing and Providing Feedback on Thinking
- Checking student work and providing insightful feedback using embedded rubrics assessing their work.
- Creating Rubrics.
- Creating Lesson Content For Students.
- Developing Lesson Sequences and Units.
- Creating Assessments Based on Learning Objectives.
- Writing Student Progress Reports Based on Collected Data.
- Creating Personalized Student Learning Plans.
- Project-Based Learning Lessons and Summative Projects.
- Developing Assessments and Differentiated Assessments
AI is Changing the World
AI is changing the way we interact with the world, each other, and how we conduct our work. There’s a large opportunity for educators to help students and our communities adjust to these technologies and navigate how they can learn how to use them in a way that benefits themselves and the community. Many new professions and jobs will arise from AI, but much will have to be done to ethically develop AI to ensure it is being used responsibly. Thus, our job as educators will be to further study and learn how we can teach this technology as it evolves over the next few years.
Another version of this Article: The ChatGPT Version of this Article – I inputted the article into ChatGPT to see if I can make it more fun and engaging to read. What do you think?
Are you curious about AI and its impact on various aspects of our lives? In this article, we will explore the five big ideas in AI literacy and how AI works. AI encompasses a variety of tools and mechanisms, such as machine learning, deep learning, neural networks, computer vision, and natural language processing, that use algorithms to analyze data and solve problems. AI is present in almost every industry today, such as banking, healthcare, and cybersecurity, where it is used to improve efficiency, productivity, and profitability. However, AI is not entirely accurate and has biases that must be filtered out through safeguards and filters that need to be put in place when data is collected and utilized by the algorithms powering AI.
AI has an impact on our daily lives in various ways, and one example is text prediction. The algorithm uses the data we provide, such as the words and phrases we use while writing, to predict the text that will follow. Another example is the predictions made for favorite websites, Netflix show preferences, and advertisements. Based on the data we have provided over time, AI can predict our preferences and even our behavior. AI can also be used to prevent fraud and cybersecurity attacks in banking through machine learning algorithms and in healthcare by using our biometric data and historical data to make predictions about our current and future health.
However, AI is only as accurate as the data it utilizes to make ongoing predictions. Therefore, the data it may have in its database may also be biased, and misinformation can easily be placed within a database that can be pulled by AI. Bias and misinformation can be easily pulled into AI, which can then be scaled to meet mass audiences (i.e., social media newsfeeds).
To filter out bias, filters and safeguards need to be put in place when the data is collected and utilized by the algorithms powering AI. We must also be judicious and skeptical of all information we see, triangulating our conclusions by synthesizing the information we process, and putting counterfactual data to filter sensitive attributes. In the end, AI literacy and its themes are essential for teaching students a variety of skills integrated with technology that has AI embedded in it.
How AI is Impacting Education
AI is having a significant impact on education. In recent years, there has been a growing interest in how AI can be used to improve the educational process. One example of this is adaptive learning, where AI algorithms are used to personalize learning experiences for individual students. By analyzing data about a student’s progress, the AI can determine their strengths and weaknesses and adapt the curriculum to meet their needs. This has the potential to improve student engagement and increase academic performance.
Another way AI is impacting education is through chatbots. Chatbots are AI-powered tools that can answer student questions, provide personalized feedback, and even grade assignments. This can help reduce the workload for teachers and increase efficiency in the classroom.
However, it is important to note that there are also potential downsides to the use of AI in education. For example, there is a risk that the use of AI may perpetuate existing biases and inequalities. Additionally, some educators are concerned that the use of AI may lead to a loss of human connection and empathy in the learning process.
Teaching AI Literacy
Given the increasing impact of AI on our lives, it is important that we teach AI literacy to students. AI literacy involves not just an understanding of how AI works, but also an understanding of its societal impact and ethical considerations.
One way to teach AI literacy is through interdisciplinary projects that incorporate AI concepts into a variety of subjects. For example, students could use machine learning algorithms to analyze data in a science project, or use natural language processing to create a chatbot in a language arts class.
Another way to teach AI literacy is through dedicated courses and workshops that focus specifically on AI concepts and their applications. This could involve hands-on coding projects, guest speakers from industry experts, and discussions of ethical considerations related to AI.
AI literacy is an essential skill for students in the 21st century. With AI playing an increasingly important role in our lives, it is important that we not only understand how it works, but also its potential impact on society. By teaching AI literacy, we can prepare students to be responsible and informed users of AI technology, and to be critical thinkers who can recognize and challenge bias and ethical issues related to AI.