Stop Asking “How Can I Use AI?” Start Asking This Instead

There’s a quiet pattern playing out in faculty meetings, prep periods, and Slack threads across higher education. Teachers are using AI to generate quizzes, draft emails, write slides, and summarize readings. None of these uses are wrong. But they share a problem that almost no one is naming.

Faster does not mean better. Efficiency is not the same as transformation.

A new article from Dr. Richard Violanti at Niagara University, published in Faculty Focus last week, calls this pattern “The Efficiency Trap.” His argument is the cleanest distillation of the AI-in-education problem I’ve read in months. Most educators are asking the wrong question. And until that question changes, AI in the classroom is going to keep delivering faster lessons of the same shape rather than fundamentally better ones.

The Efficiency Trap

Violanti’s framing is sharp. When AI enters your classroom as a productivity tool, the workflow changes but the instructional model does not. You used to write quizzes by hand. Now AI writes them in seconds. The quiz is the same. The student experience is the same. The learning is the same.

This is not a new pattern. Decades of research on educational technology show the same thing. Innovations enter classrooms, get absorbed by existing practices, and quietly leave learning unchanged. The smartboard replaced the chalkboard but did not change what was taught. The clicker replaced the show of hands but did not change how students were assessed. AI is on the same trajectory unless we ask the better question.

The Better Question

Violanti reframes the conversation with one substitution. Instead of asking what AI can do for you, ask what AI can do for the student-centered learning experience you are trying to build.

The first question puts the educator at the center. AI becomes a tool that saves you time. The second question puts the student at the center. AI becomes a tool that changes what learning looks like for them.

That difference is structural, not semantic. It changes which AI features matter, which assignments survive, which guardrails you need, and which ones you can let go of. Once you make the shift, you stop chasing tools and you start designing experiences.

This Is The Seed

If you’ve been around Harvest Kernel for any length of time, you know that the SeedStacking™ methodology starts with the same move. The Seed is not the tool. The Seed is the question. Get the question right and the right tool stack reveals itself. Get the question wrong and you spend months churning through apps, plugins, and platforms that never actually help.

The three practices below are tool agnostic. They work whether you’re using Gemini, ChatGPT, Claude, NotebookLM, or whatever launches next month. That’s the point. Frameworks last longer than tools.

Three Practices That Keep Students at the Center

1. Use AI to Differentiate, Not Just Generate

The classroom you actually teach is heterogeneous. Different reading levels, different prior knowledge, different language backgrounds, different attention patterns. The “one reading for everyone” model has always been a compromise driven by what was practical, not by what was best for students.

AI changes the math. With minimal effort, you can take the same scholarly reading and offer it as text leveled at three reading abilities, an audio podcast summary, a short video walkthrough, vocabulary supports embedded inline, and alternative explanations that approach the concept from different angles. Students choose what helps them learn. You didn’t lower the bar. You met more students at theirs.

Try this: pick one reading per week. Generate two or three alternative versions with AI. Make all of them options. Watch what students actually choose. The result is not just efficiency. It’s increased access and equity.

2. Turn AI into a Thinking Partner

This is where most AI policies break. The fear is that students will use AI to produce final answers and stop thinking. That fear is reasonable. It is also avoidable. The difference is in assignment design.

When a student uses AI to “write the essay,” learning collapses. When a student uses AI as a sparring partner, learning deepens. Violanti offers four examples that any educator can adapt:

• History students interview an AI persona of a historical figure and have to identify where the AI’s responses are anachronistic or contextually off
• Business students debate inside an AI-generated ethical dilemma where the AI plays the opposing case
• Education students analyze AI-generated classroom case studies and decide how they’d intervene
• Science students stress-test their own hypotheses by asking an AI lab assistant to find the flaw in their reasoning

The structural addition: require a short student reflection that names what the AI got right, what it got wrong or oversimplified, and how the interaction changed their thinking. Now AI is a catalyst for critical thinking, not a shortcut around it.

3. Use AI to Design Better Learning Experiences

Most instructors use AI to generate materials. Fewer use it to design experiences. This is where the real leverage lives.

Instead of “Write me a quiz on photosynthesis,” try “Give me three different ways students could explore photosynthesis through discussion, collaboration, or problem solving, calibrated for 9th grade biology in a 50 minute period.” Now you’re using AI for instructional design, not content production. The output is a menu of pedagogical choices, not a finished worksheet.

Adapt what you get. AI doesn’t know your students or your curriculum the way you do. But it can expand the option space when you’re tired and stuck and out of ideas at 9 PM on a Sunday. That alone is worth the price of admission.

The Takeaway

Where to Go From Here

Pick one of the three practices above. Try it next week. With one class. With one assignment. The smallest possible test.

That’s the SeedStacking™ rhythm. Plant one Seed. Watch it Sprout. Decide whether to Grow it. Then Harvest what you’ve learned and stack the next one. Sustainable change in teaching practice has never come from sweeping overhauls. It comes from one deliberate experiment at a time.

Violanti’s full article is worth reading. Read it here at Faculty Focus. Bring it to your next department meeting. The conversation it starts is more valuable than the next AI tool launch.

Plant ideas. Cultivate skills. Harvest results.