For decades, artificial intelligence (AI) has powered intelligent tutoring systems, translation tools, voice assistants, and more. However, the rise of generative AI introduced new urgency—and unevenness—in K–12 education.
Some schools immediately restricted access to AI tools, citing legitimate concerns about cheating, student privacy, bias, and cybersecurity. Meanwhile, others embraced experimentation, identifying opportunities for increased efficiency and automation. Current guidance has largely felt reactive and overly restrictive, leading to fragmented implementation shaped by market availability rather than educational values.
If AI is to play a meaningful role in advancing teaching and learning, the field needs more than policies that manage risk or frameworks that encourage experimentation. While absolutely necessary, these things alone are largely insufficient. Instead, the K-12 sector needs a shared strategy grounded in learning, centered on students, and designed to help systems move from isolated pilots to sustainable, coherent practice.
If AI is to play a meaningful role in advancing teaching and learning, the field needs more than policies that manage risk or frameworks that encourage experimentation.
The emergence of AI offers a chance to rethink the instructional core: what students learn, how they learn it, and the systems that support them. The field needs a compelling vision for how AI can help education systems do something fundamentally different. The challenge is to bridge the gap between bold ambitions and day-to-day reality. FullScale (formerly The Learning Accelerator and the Aurora Institute) has started to take on this task.
Through our work with educators, leaders, researchers, and policymakers, we have identified five essential shifts for building a more inclusive, responsive, and effective K-12 education system.
| #1: Personalized Learning
Personalized learning at scale has been a goal for education leaders since digital technology entered classrooms. Despite this promise, the focus has often been on the delivery and pacing of content, reducing personalization to an association with self-paced, digital content instead of a broader, student-centered approach to instruction. AI offers the potential to overcome the persistent barriers to scaling this vision– limited time, capacity, and real-time insight. The intention is not to replace teachers with screens, but to pursue a model where personalization is grounded in strong relationships and supported by timely insight. AI promises to make relevant practices such as formative assessment and flexible grouping more sustainable by reducing logistical burdens. For example, as we are learning through our Accelerating Adoption Network, ASU Preparatory Academy is scaling Archie, an AI-powered tool that delivers personalized, real-time feedback to help students better understand Algebra concepts. As students use AI to reflect on their progress and learning, teachers receive real-time insights. Archie then enables timely, targeted teacher support, increasing opportunities for personalization and making responsive learning more sustainable. #2: Mastery-Based and Competency-Aligned Progressions Educators have long sought systematic, sustainable structures that allow them to track when, how, and to what degree students demonstrate their mastery of content and skills. AI has the potential to support this work at scale, solving the logistical issues of tracking student progress, adapting pacing, and assessing skills that don’t always fit into traditional systems. From curating portfolios to surfacing patterns that inform real-time instructional adjustments, AI can extend teachers’ capacity to focus on student growth, ensuring that students move through content when ready rather than when a calendar dictates. Additionally, AI has the potential to support the creation of competency continua (developmental trajectories in which each new learning opportunity builds on and is informed by the one before it), freeing up precious time for educators by giving them a head start on the development of continua and progressions to build from. More importantly, AI may help educators recognize and value learning that has historically gone under-measured. Tools such as those found in Building21’s Beacon platform help educators and students focus on progress, reflection, and performance. The AI Project Builder provides step-by-step guidance to help students design personalized projects aligned to specific competencies and standards, and the AI Feedback Tool delivers immediate, actionable insights to support skill development. By illuminating students’ approaches to inquiry, reflection, and problem-solving, the AI provides a more complete picture of not just what students know, but how they learned it. #3: Flexible Learning Environments Fixed schedules, physical classrooms, and time-based pacing – all hallmarks of traditional K-12 systems, rarely reflect the complexity of students’ lives. AI presents an opportunity to design learning environments that are more flexible, connected, and responsive, as students navigate between in-person, virtual, independent, and community-based experiences. By reducing logistical burdens, AI can serve as a scaffolding mechanism, managing complex logistics, surfacing insights, and streamlining coordination without displacing the human elements. Consider Da Vinci Connect TK-8, a public, hybrid homeschool in California, where students engage in a blend of in-person instruction, independent work at home, and interdisciplinary, project-based learning. This model expands where, how, and with whom learning happens, as teachers coordinate shifting schedules, adapt projects to individual students, and ensure continuity across contexts. While Da Vinci Connect is not yet extensively leveraging AI, with thoughtful integration, it could support communication, coordination, and instructional responsiveness without replacing the human relationships at the heart of its model. #4: Access to Meaningful Learning Students need more than access to information and AI-enabled feedback; they need learning experiences that help them build understanding to make sense of the world and apply knowledge in meaningful ways. These experiences foster academic knowledge as well as durable skills such as curiosity, reflection, and the ability to transfer learning across contexts. When used intentionally, AI can help educators differentiate content, scaffold inquiry, surface content that reflects multiple perspectives, pose generative questions, and encourage students to synthesize ideas. By centering inquiry and reflection, Coursemojo – another Accelerating Adoption Network participant – makes grade-level reading and writing tasks highly interactive and differentiated. Within the curriculum-aligned, AI-powered platform, students receive real-time feedback and scaffolded questions. Teachers then access a dashboard showing every student’s level of understanding. These supports allow teachers to better differentiate their instruction while students connect ideas and refine their thinking. #5: Student Agency Students learn best when they have agency, own their learning, set meaningful goals, reflect on their progress, and choose how to engage. AI has the potential to support agency when it functions as a co-pilot that elevates student thinking, supports metacognition, and surfaces insights, all when students choose to use it. To enable student agency at scale, instead of more AI-powered tools, schools and systems need learning models and frameworks. Dr. Sabba Quidwai developed the SPARK Framework to guide learners and educators through a thinking routine. It asks them to describe the Situation; identify the Problem; articulate their Aspiration, define measurable Results, and share Kismet or a surprise. SPARK provides a thinking routine rather than a list of questions or prompts. Instead of replacing student thinking, it encourages them to leverage AI as a thought partner so that they engage with it more purposefully, ensuring they lead their own learning with clarity and intention. |
Bridging the Gap from Urgent Reality to Audacious Future
This moment is about more than managing risk. It’s about designing toward possibility. The five essential shifts identified above can lay the foundation for a more student-centered and future-ready education system. To help schools, systems, and states make hops, skips, and leaps towards the future, FullScale has designed a series of resources to move the field from fragmented responses to a coherent strategy for transforming education with AI. Together, they offer guiding principles, actionable tools, and policy recommendations to ensure AI supports powerful teaching, meaningful learning, and expanded opportunities for all students.
© Cognia Inc.
This article may be republished or reproduced in accordance with The Source Copyright Policy.
The information in this article is given to the reader with the understanding that neither the author nor Cognia is in engaged in rendering any legal or business advice to the user or general public. The views, thoughts, and opinions expressed in this article belong solely to the author(s), and do not necessarily reflect the official policy or position of Cognia, the author’s employer, organization, or other group or individual.
