AI Literacy in the Classroom
What should we be teaching students about AI?
How do we prepare them to use this tool effectively as a digital media professional?
See what Ellia Morse, Founder of Authentiya, recommends to teachers who struggle with these challenges in the classroom. Learn about grants currently available to support adding AI literacy to your curriculum.
My name is Ellia Morse and I’m the founder of Authentiya which promotes AI literacy in the classroom. I’m a recent KU graduate (rock chalk!), software engineer, and I serve on the USD 497’s Generative AI and Emerging Technology advisory board. I wanted to share some of what I’ve learned about AI in my journey of building Authentiya and promote our micro-grant program! As a new start-up backed by the KU Innovation Center and the ACCEL-KS program, we’re looking to support teachers with funding and training as you prepare your students to use AI responsibly.
Some of you may have seen the recent article where a college professor created a trap to catch his students using AI. He included, in white font, the word “Madagascar” expecting that students wouldn’t see it but whatever AI they copied the instructions into would. Thirty-two of his thirty-five students failed portions of the midterm for using AI.
In this game of cat and mouse where teachers leave traps for students using AI while students attempt to humanize AI output and add back typos, classrooms overlook the fact that AI is becoming necessary for career readiness. The National Association of Colleges and Employers says that one third of entry-level positions are expecting AI literacy. Sixty percent of workplaces said that they are assigning interns projects that require AI tools. Schools need to prepare students with AI literacy but progress is disjointed as most classrooms don’t have formal policies around AI usage, ban it altogether, or hope that another teacher has already covered it.
Here are some key topics I'd recommend discussing with your students:
How AI is trained: Many students don’t understand how AI content is generated and just view the prompting box and output without context. They do not know that AI was trained on multitudes of other creators' work, or that companies that produce AI text, pictures, video, or music are profiting from artists that did not give permission for their work to be trained on.
AI and Creativity: There’s an ongoing debate of whether it’s okay for AI to get “inspiration” from those artists or whether it’s directly stealing. Regardless, there have been large backlashes due to AI usage. One of the writers of Kpop Demon Hunters admitted to using AI when writing the song “Soda Pop” to negative fan reception. However, people were more understanding when they learned that it was just used during the ideation phase and “gave [the writer] ideas to make the song more bubbly” rather than writing the song wholesale. Another example is Clair Obscur: Expedition 33 being stripped of its Game of the Year award by the Indie game awards after an AI-generated placeholder texture was found in the game. Another consideration is that works created entirely by artificial intelligence without meaningful human creative input cannot be copyrighted. Creatives need to understand where AI can be used as a partner rather than a replacement for their own work.
Environmental impact: AI has a large environmental impact but it can be difficult to pin down figures. AI water usage notoriously fluctuates between reports. Google says a query uses five drops of water, while Sam Altman of OpenAI has stated 1/15th of a teaspoon. These numbers both ignore the amount of water that is used to train AI models before it even reaches final users, estimated to be around 50% of total water usage. These numbers are muddied between what water comes from local grids which provide the municipal water that comes from our sinks and sprinklers, and “blue” water sources such as lakes, oceans, and rivers. CBS gives a point of perspective, saying that in America, data centers use half as much water as golf course irrigation does. Some folks might call both a waste of water but it’s important for students to have a sense of perspective.
Bias: Since AI is trained by example, if AI sees a lot of one thing it will over amplify it in its output. For example, OpenAI acknowledged that its models since GPT-5.1 had an odd quirk where it fixated on goblins. This is because an oddly large amount of its training data mentioned goblins and gremlins. This plays out in less benign ways in day-to-day output. If a model is asked to produce a picture of a doctor, you may overwhelmingly see a picture of a white man even though men make up only about 60% of physicians in the United States. Researchers at Stanford found that hiring tools using AI frequently discriminated against Black and Asian applicants, leading to systemic rejection.
Hallucinations & Misinformation: AI can often make up information that doesn’t exist because it doesn’t truly “know” information the way we do. AI functions by predicting the next word in a sentence, pixel in a piece of art, sound in music. When asked to cite quotes in a book, AI might invent a passage just because it sounds like something that could plausibly exist in that story. You might have seen a video of raccoons jumping on a trampoline, which was shared without many people realizing that it was AI-generated. When people blindly trust AI output, they can unknowingly become propagators of misinformation. Students need to understand that they hold responsibility for any AI output they share. Remind them to not share anything that they don’t want to represent them.
Emotional impact: A report from Harvard said that one in four Gen Z young adults are using AI as a friend. AI offers a frictionless relationship where models appear as a friend with a human personality. I have had several conversations with people who have said “I was talking to ChatGPT and he told me…” using human pronouns rather than treating AI like the inanimate object it is. Even though 80% of this audience surveyed said that they believe AI makes you lazier, these conversations are addictive. Audiences keep returning as they start to depend on it. During a classroom visitation, I asked one high school freshman why they used AI to write an assignment and they said “Why would I try when AI can just do so much better than I could?”
I would highly encourage you to spend a day of your curriculum discussing some of these topics with your students. We want to equip students to think about AI critically and to use it as a tool rather than a crutch.
As you consider how to incorporate AI literacy training into your curriculum, I would love for you to reach out! We are currently offering micro-grants through September 15th to support teachers as they educate their classrooms about AI. We’re running a free pilot program this semester with training modules to allow your students to “Earn their AI driver's license”. We also provide software for students to use AI while writing and ideating while providing you with transparency into exactly how they use it on assignments. If you’re not sure whether it is a good fit or want to hear more, I’m happy to chat! You can reach me at ellia@authentiya.com.