Teaching & Learning in an Age of AI
As I begin working on my 2026 programming that will focus on teaching and learning in an age of AI (join our email list for updates), I’ve been doing deep dives into AI topics of the greatest interest to educators and instructional designers across sectors. One exploration approach I’ve been trying is to engage ChatGPT, Claude, Copilot, Gemini, and Perplexity in a duel of analyses on the same prompt, then have Notebook LM take a crack at synthesis.
I’ve been trying this approach for some time now as I designed our AI Pathfinder Foundations course (see below), and I’ve found it to be an interesting way to brainstorm. However, it doesn’t come anywhere near replicating an academic literature review. Yet, the results were interesting enough that I thought that I’d share them here as I think the themes and challenges are worthy of further investigation.
Unfortunately, it’s impossible to pin down sourcing for specific “facts” in this type of approach without devoting substantial time to fact-checking. Also, in much of the analysis, regular old “educational technology” could serve as a placeholder for AI. So, I’ve decided to reprint the Notebook LLM analysis in its entirety, add some annotations [in brackets] with light commentary and questions, and identify where the most glaring citations are missing but needed.
Interested to hear any thoughts in the comments!
The AI Revolution in the Classroom: Understanding the Top Challenges for Educators
1. Introduction: A New Era for Education
Artificial Intelligence (AI) is no longer a concept from the distant future; it is a present-day reality in our schools, colleges, and workplaces. The rapid rise of powerful generative AI tools has created a critical inflection point for education, offering exciting possibilities for personalized learning and efficiency. However, this new era also presents significant challenges that educators must navigate.
This document [prepared by a Notebook LLM based on its analysis of reports from ChatGPT, Claude, Copilot, Gemini, and Perplexity] explores the most pressing problems educators face today. We will delve into the crisis of academic integrity, the urgent need to redesign assignments, and the fundamental importance of understanding what AI is, how it works, and how to use it ethically. While the promise of AI is great, its successful integration depends on thoughtfully addressing these hurdles. We begin with the most immediate and widespread challenge: ensuring student honesty in a world where answers are just a prompt away.
2. The Integrity Crisis: A New Definition of “Cheating”
The widespread availability of AI has triggered a crisis in academic integrity. For educators, the core pain point is the struggle to determine if student work is authentic. One analysis reveals that 71% of teachers struggle to distinguish AI-generated work from a student’s own efforts. [citation?]
This uncertainty has created a difficult “cat-and-mouse” dynamic, where educators feel they are “policing” students instead of teaching and mentoring them. The situation is made worse by traditional AI detection tools, which are often unreliable and have high false-positive rates. An inaccurate accusation of cheating can severely damage the trust between a student and teacher, turning the classroom into an adversarial environment. This breakdown of trust forces a critical question: instead of trying to catch students, can we change the nature of schoolwork itself? This question is impossible for individual educators to answer in a vacuum, which is why the integrity crisis is so deeply connected to the absence of clear institutional rules.
3. Rethinking the Test: The Rise of “Authentic Assessment”
To escape the adversarial cycle of policing and detection, educators must make a decisive pedagogical shift toward authentic assessment. [Why limit to “assessment”? What about the rest of the learning experience?] This is not merely an alternative; it is the essential strategy for restoring integrity and relevance to student evaluation. In simple terms, authentic assessments are tasks [Ah, so this is characterising an “assessment” as the learning task?] that require students to demonstrate their thinking, not just their access to information. By focusing on process, real-world application, and personal synthesis, they create an evaluation environment where AI is a tool for work, not a substitute for it.
Key features of authentic assessments include:
- Focus on Process, Not Just the Final Product: Educators evaluate a student’s entire learning journey—including drafts, reflections, and decision-making logs. This makes it difficult to simply copy and paste a final answer from an AI, as the process of thinking becomes a visible and gradable part of the work.
- Require Higher-Order Thinking: Assignments [oh, good! We moved on from “just” assessments”] are designed to demand skills like critical thinking, creativity, and context-specific problem-solving. Examples include oral defenses of a project or developing solutions for real-world scenarios. To be effective, these scenarios must be accessible and culturally relevant to all students, preventing the assessment itself from reinforcing the “AI divide.”
- Use AI as a Transparent Tool: Rather than banning AI, some authentic assessments invite students to use it as a tool and then demonstrate their own intellect by critiquing, refining, or building upon the AI’s output. This approach allows educators to assess a student’s ability to think critically with technology, a vital skill for the modern world.
To design these innovative assessments [Oh, we’re back to “assessments”] , however, educators must first develop a clear understanding of the technology they are working with.
4. The Knowledge Gap: What is “AI Literacy”?
AI Literacy [definition?] is no longer an elective or a niche technical skill; it is a foundational competency for modern educators and students. It is the critical lens through which all other AI-related challenges—from integrity to equity—must be viewed. It is not about becoming a coder or data scientist; it is about becoming an informed and critical user of technology who understands AI’s capabilities, limitations, and potential for bias.
Unfortunately, a significant knowledge gap exists for both educators and students:
- Approximately 40% of faculty report that they are just beginning their own AI literacy journey. [citation?]
- An overwhelming 80% or more of students say their teachers have never taught them how to use AI tools appropriately for schoolwork.[citation?]
The goal of AI literacy is to cultivate an informed, “skeptical-optimistic mindset.” [citation?] This balanced perspective allows both teachers and students to use AI as a powerful assistant without blindly trusting its output, preparing them to navigate a world where AI is ubiquitous. Building this individual literacy is the first step, but it must be supported by clear, institution-wide rules.
5. The Rules Vacuum: The Urgent Need for Clear Policies
The absence of clear institutional AI policy has created a leadership vacuum, forcing educators into the untenable position of acting as individual arbiters of ethics and integrity. This ad-hoc approach is unsustainable and presents a significant risk to institutions. Without guidance, a student might be praised for using AI in one class and punished for it in another.
Educators report feeling like they are “flying blind,” forced to make ad-hoc decisions about what constitutes cheating. [citation?] Data confirms [confirm] this reality, with over 80% of schools lacking clear and comprehensive AI policies. [citation?] This absence of top-down guidance leaves teachers feeling unsupported and vulnerable as they navigate complex ethical situations alone. Just as rules must be applied fairly to all students, access to the technology itself must also be equitable.
6. The Fairness Question: Bridging the “AI Divide” [extension of “digital divide”]
Beyond clear rules, the integration of AI into education raises critical questions of fairness and equity. Two major concerns stand out:
- The “AI Divide”: There is a growing gap between students who have access to premium AI tools, high-speed internet, and quality training, and those in under-resourced schools who do not. This divide threatens to systematically disadvantage students who lack the infrastructure and support to build AI fluency, worsening existing educational inequalities.
- Algorithmic Bias: AI models are trained on vast amounts of data from the internet, which contains human biases. There is a significant risk that these tools can have built-in biases that reinforce stereotypes or produce content that is unfair to certain student populations.
Addressing these issues is essential for ensuring that AI is implemented in a safe, fair, and inclusive way. These challenges not only affect students but also force educators to reconsider their own professional identity.
7. The Future of Teaching: A New Role for Educators
Many educators feel anxious about how AI will change their professional roles, with some wondering if technology could eventually replace them. [extends technology integration concerns of yore] The goal is not for teachers to compete with AI, but to cede automatable tasks to the machine and double down on the deeply human work that technology cannot touch: mentoring, coaching, and inspiring. [sounding VERY familiar] The consensus among experts [citation?] is that AI will not replace teachers, but rather transform their roles by shifting the educator’s focus from a “content deliverer” to a “learning facilitator, mentor, and designer.” [sounding VERY VERY familiar]
This transformation allows teachers to focus on the uniquely human strengths that AI cannot replicate.
| Traditional Focus | New AI-Era Focus |
| Delivering information and content. | Mentoring and providing social-emotional support. |
| Spending hours on administrative tasks like grading. | Facilitating complex problem-solving and critical thinking. |
| Feeling disconnected from students due to technology. | Using AI-driven time savings for more one-on-one student engagement. |
[sounding VERY VERY VERY familiar]
Research and educator reports show that AI can save anywhere from 3 to 10 hours per week on average by automating administrative tasks. [CITATION?] The goal is to reinvest this recovered time directly into the “human work of teaching,” strengthening the relationships and deep learning that lie at the heart of education.
8. Conclusion: Beyond the Tool
Artificial intelligence is far more than just a new classroom tool; it is a powerful catalyst forcing a fundamental re-evaluation of education itself. The challenges it presents—from upholding academic integrity to ensuring equity and redefining the educator’s role—are pushing schools and colleges to innovate in ways they never have before. The path forward requires a “skeptical optimistic” mindset, one that acknowledges both the risks and the rewards. The ultimate goal is not to resist AI, but to guide its use thoughtfully and ethically to create a future where technology serves a more engaging, equitable, and human-centered vision of learning.
As a reminder: Join our new course!
In case you missed it, our AI Pathfinder Foundations self-paced course and private online community are officially open for enrollment with an awesome roster of early participants! This beginner-focused learning experience is designed to help build the skills to explore five popular AI Assistants, including ChatGPT, Claude, Gemini, Copilot, and Perplexity, using frameworks for evaluating, using, and verifying the output of any conversational AI.
Check out answers to common questions in our FAQ on the course overview page, or reach out via email at: [email protected]
Thank you for following along. I’m excited to explore AI alongside you.
– Jennifer
Want to follow along as I document this journey?
I’ll be sharing Next Path Notes about building this course, my AI discoveries, and behind-the-scenes insights from launching a new venture.