As a school leader who uses AI tools every day, I can say that AI is great – and terrifying at the same time. It has been a few years since ChatGPT rolled onto the scene, yet we are still very much in the pioneering stage, especially in education. Understanding the contrast and recognizing that children and educators have remarkably imperfect productivity tools at their disposal is an important first step in engaging with AI tools. Balancing that, while recognizing time tested teaching and learning practices remain a part of our new AI world.
Important non-AI practices remain for educators and students. A spectrum of analog methods must be part of teaching and learning. In fact, reinforcing that these practices are more important than ever is crucial. Researchers are exploring ways that this balance can perpetuate the humans/AI connection, working together.
Consider that sometimes, a back-to-basics approach works best. For instance, play has long been documented as an effective and engaging way to learn. A meta-analysis confirms that analog games remain a resilient, adaptable form of active learning that complements technological innovation by fostering conversation, debate, laughter and interaction among real people.
Implementing inclusive classroom practices is one of the most effective ways to design AI-adjacent teaching and learning approaches. Students frequently turn to AI when they feel overwhelmed, confused or lack the confidence to begin an assignment. By integrating supportive frameworks directly into daily classroom practices, educators can build student confidence and foster genuine learning.
Here are some inclusive, Alternative to AI approaches to consider:
- Structured brainstorming templates: Offer graphic organizers and planning tools. This inclusive practice provides a clear, accessible entry point for students who struggle to write from a blank page or screen.
- Process-focused feedback to students: Incorporate regular check-ins where students identify their drafting plans and hurdles. To maintain an inclusive environment, allow students to use other methods, such as speech-to-text tools, to draft these updates.
- Scaffolded assignments: Break larger projects into smaller, sequential steps with built-in checkpoints. This embeds the learning process directly into the classroom timeline.
- Exemplars and rubrics: One of the best ways teachers can give students a reference point is to provide a model project developed by a prior student. Provide clear examples and explicit rubrics to model expectations, removing ambiguity and supporting diverse learning needs.
While these practices are effective for many learning opportunities, the reality is that AI is here to stay and can and should be used by educators and students.
Consider also how students can be guided to make effective use of AI as a learning partner:
Shift: from “cheating tool” to “study buddy”
When students treat AI as a tool, they often use it passively to generate quick answers, which stifles actual learning. But when they learn to treat it as a learning partner, the dynamic changes. They can bounce ideas off of, break down confusing concepts, or get feedback on their writing. With this approach, the student still does the heavy lifting, but the AI acts as a collaborative co-pilot.
Cautions to consider:
- Over-reliance: Students who blindly trust AI to do their thinking experience a drop in actual skill development. Their stress also spikes when they face in-class, screen-free assessments because they’ve lost their crutch.
- Total distrust: On the flip side, completely banning AI or refusing to engage with it creates a different kind of stress. Students and educators miss out on powerful, personalized learning aids and risk falling behind in digital literacy.
Tools like Gemini’s guided learning, helps in areas like math, but may struggle in writing support, or as this tester remarked, was messier in other areas. This reinforces the need for analog and AI tools in teaching and learning. Good for some, not all.
The goal: Teaching calibrated trust
For AI to work in education, lesson design needs to consider AI literacy. Students must be taught exactly where AI shines (like organizing random brainstorm notes or explaining a complex theory in simpler terms) and where it fails miserably (like inventing facts, biased logic, or lacking genuine human nuance).
AI fails with human nuance!
AI outputs look somewhat impressive on the surface, but upon deeper inspection, they produce far less qualitative content. An AI cannot replace the relational, adaptive and emotionally intelligent aspects of human teaching. While AI can efficiently deliver facts, it fails to read between the lines of student behavior, messy drafts and diverse cultural backgrounds. In fact, there is still a strong concern about bias in AI outputs, as AI accesses an internet ecosystem that remains biased.
AI naturally flattens the learning process because algorithms lack human empathy. The case for partnering with AI remains, and balancing that partnership must be a priority in teaching. Human learning can be structured around analog and digital methods, not to reduce but to increase learning capacity by identifying and taking the best parts of each.
Academic success in the modern classroom doesn’t come from outlawing AI or surrendering to it. It comes from teaching students how to manage the partnership — knowing when to lean on the technology and when to trust their own minds.
School communities should teach AI literacy alongside critical thinking, emphasizing that AI is a supporting tool rather than a replacement for human creativity and judgment. Establishing clear guidelines will help students and educators actively balance screen-based AI assistance with hands-on, analog methods to ensure original thought and genuine understanding.
Opinions expressed by SmartBrief contributors are their own.
Subscribe to SmartBrief’s FREE email newsletters to see the latest hot topics on educational leadership in ASCD and ASCDLeaders. They’re among SmartBrief’s more than 200 industry-focused newsletters.