The AI-in-Education Year-in-Review: 5 Lessons We Learned in 2025
Five lessons from 2025 that show how AI in education moved from hype to strategy — and what districts should carry forward into 2026.
If 2023 was the year artificial intelligence (AI) arrived in schools, and 2024 was the year everyone tried it, 2025 was the year education finally got serious about using it well. AI seems to be changing daily, and it can be hard to keep up with. Still, this year was all about teachers, tech directors, and district administrators finding ways to incorporate AI safely and effectively into their schools and classrooms. These 5 lessons show how far we’ve come in 2025 and help us plan for 2026.
Lesson 1: AI Moved from Hype to Strategy
Early AI adoption was a lot like the Wild West–everyone was just doing whatever they wanted, with whatever tools they chose to use. It was chaos! In 2025, districts started to get smart about their AI usage, asking questions like, “Where does AI fit?” and “What problems does it solve?”
Many districts around the country formed AI committees and working groups. They had conversations with stakeholders. They created policies and strategic plans. They put pedagogy first, then figured out ways to augment that with AI. (Check out the great work being done by the Ottawa Catholic School Board and Buffalo Public Schools!)
AI is no longer just a “cool” fad.
Lesson 2: Districts Began Building the Foundations—Policies & AI Literacy
“We just ban all use of AI.” A common statement in 2024, but no longer acceptable in 2025. Bans don’t work, and they don’t allow teachers and students to develop the tremendously important skill of AI literacy.
Digital Promise defines AI literacy as: “The knowledge and skills that enable humans to critically understand, evaluate, and use AI systems and tools to safely and ethically participate in an increasingly digital world.” Our students will face a world filled with AI tools–they must be empowered to thrive in this environment. Teachers can only guide their students in this learning journey if they are also AI literate.
To this end, districts started to consider how to best incorporate AI tools into their ecosystems. AI policies that have been developed outline expectations for teachers and students, allowing them to be informed users of AI. Data privacy and proper vetting of tools were critical elements this year as districts moved forward with AI usage.
As districts focused on building AI literacy and thoughtful policies, another realization quickly emerged: the real challenges weren’t about the tools themselves, but about how students understood and used them.
Lesson 3: “Student Misuse” Isn’t an AI Problem—It’s a Literacy Problem
A lot of early talk about AI involved how students would use AI to cheat or avoid tasks that involve deep thinking–the same concerns that were raised when calculators were made widely available and when students could look things up on Google.
While academic dishonesty and cognitive offloading are concerns, the underlying problem is that students need to develop AI literacy skills. Students generally want to learn and want to do the right thing. AI misuse, then, is often due to a lack of skills, not malicious intent.
To support students, districts need clear guidelines about when AI use is appropriate and when it is not. In class, teachers should help students learn how AI works, ways AI can support their thinking and learning, and how to evaluate outputs for accuracy and bias. (Does this sound like too much for teachers? This webinar from Lindy Hockenbary demonstrates that AI literacy can easily be incorporated into instruction, without adding more onto teachers’ already full plates.)
Helping students use AI responsibly was only part of the story. The next challenge was addressing a broader concern: what AI meant for the role of teachers themselves.
Lesson 4: AI Supported, Not Replaced, High-Quality Teaching
One of the early fears swirling around in the AI-in-education space was that AI had the potential to replace teachers. As schools and districts started to take a more measured approach to AI integration in 2025, it became clear that teachers aren’t going anywhere! AI is a tool, like a hammer - useful in some situations, but not a replacement.
AI can help teachers with all sorts of things. It can write lesson plans, provide feedback, and differentiate materials. It can streamline a teacher’s administrative tasks to give them the precious gift of time.
But AI can’t read a student’s body language and recognize that they are having a bad day. AI can’t create relationships with students (not real ones, anyway!). AI can’t inspire students the way a great teacher can. The human element is critical to education, so there will always be a need for excellent educators.
To extend the hammer analogy, a good carpenter knows when it is appropriate to use a hammer. He also knows that a hammer isn’t the right tool when he wants to tighten a screw. Likewise, AI is appropriate to use in some situations, but other tools are better suited in other situations. A successful teacher in the age of AI will know the difference and be able to use AI well when it supports instruction.
AI will never replace teachers, but teachers who are able to use AI well will replace those who can’t. Which brings us to Lesson 5…
Lesson 5: The Need for Continuous PD Became Obvious
AI is a powerful tool. It’s also ever-evolving. In 2025, districts started to realize that one-off workshops were helpful, but not enough. Teachers walked out of introductory sessions excited about what AI could do, but unsure how to turn those ideas into day-to-day practice. With already overwhelming workloads, very few have the time or mental space to experiment on their own—and without follow-up, there’s no opportunity to ask questions, get feedback, or troubleshoot.
Again and again, educators expressed a need for structured support: practical examples tied to their grade level or subject area, time to practice, and opportunities to revisit the work once they had tried AI with students. District leaders echoed the same reality in their planning conversations throughout the year: We need a clear plan for how we can support teachers.
That shift in mindset is the real story. 2025 was the year districts realized that AI integration isn’t about a single workshop—it’s about building sustained capacity. And with AI tools evolving so quickly, ongoing PD cycles, coaching, and embedded support aren’t “nice to have” anymore. They’re becoming the foundation for safe, meaningful, and equitable use.
As teachers began building that capacity, one thing became clear: the tools they use matter. 2025 brought countless new options, but only a handful genuinely supported teacher workflow and student learning.
Bonus Lesson: A Few Tools Truly Delivered
In 2025, teachers were bombarded with AI tools–a handful demonstrated real staying power. Here are four that are game-changing for teaching and learning:
Brisk Teaching — Brisk Teaching is the multi-purpose tool of AI in education. It can be used to create literally any instructional material for the classroom (from scratch or from existing documents, websites, or YouTube videos), provide feedback to students, change the language or reading level of a text, and so much more!
Snorkl — Snorkl burst onto the scene as a fantastic option for providing instant feedback to students on a variety of inputs (voice recording, whiteboard drawings, typed text). Students respond to a question in the way they feel comfortable, they receive instant feedback, and they are offered the opportunity to try again and improve their work.
Canva — Canva was already a powerful tool in education. With the addition of the AI image generator and Canva Code, the possibilities are nearly endless. Educators can create interactive online learning experiences for students without needing to know how to code.
NotebookLM — NotebookLM is an AI-powered tool that just keeps getting better. Users (teachers or students) can input documents into a notebook, such as all of the materials for a science unit. Then, the user can do all sorts of things like ask questions about the material, create a podcast or video summary, make flashcards relevant to the information, or, in a recent addition, create infographics based on the text. All of the answers or outputs are based on the provided documents. This makes NotebookLM a powerful study tool.
Where We Go in 2026
2025 laid the foundations—strategically, ethically, instructionally.
2026 needs to be the year of:
deeper AI literacy for students and teachers
expanded student use
stronger PD structures
equity-driven implementation
and alignment with district goals
We’ve moved past hype. We’re ready for purpose. So the question now isn't whether your district will integrate AI in 2026—it's whether you'll do it thoughtfully or reactively.
If your district would like to approach AI integration thoughtfully in 2026, I’d love to collaborate. I’m supporting districts with policy development, AI literacy, and practical PD structures—reach out if you’d like to explore what that could look like for your team.