Six Territories for Disciplinary AI Literacy
Not another set of guidelines, but a map of the intellectual territory we must traverse together.
Last week marked a turning point in the work I’ve been doing around disciplinary AI literacy.
On Friday, I sat with district leaders from across Central Ohio designing a workshop series built around a radical but simple premise: AI literacy must be disciplinary-specific, embedded within subject-area instruction where students simultaneously build content knowledge and critical AI capabilities. Teachers will learn to position AI as an interactive source that students interrogate using the critical thinking practices already central to their discipline.
On Monday, I led a roundtable at the Ohio School Boards Association conference with over 200 educators. The energy was extraordinary. These were people determined to prepare students for AI engagement that maintains intellectual agency while driving genuine academic outcomes.
Meanwhile, over 30 educators—university faculty and K-12 practitioners—have reached out asking to join a DSAIL research cohort focused on disciplinary AI literacy. We’re capping this first session at those initial 30, but I’ll use this newsletter to report out on what we’re learning.
These experiences have crystallized a comprehensive framework for thinking about disciplinary AI literacy across K-16 education. Not another set of guidelines, but a map of the intellectual territory we must traverse together.
Territory 1: Epistemology and Disciplinary Ways of Knowing
AI fundamentally disrupts the relationship between process and product in disciplinary work. Students can now obtain sophisticated outputs without engaging the processes that generate disciplinary knowledge.
A student can receive a well-structured historical argument without learning to evaluate sources. They can get a correct mathematical solution without developing proof logic. They can produce literary analysis without learning close reading.
This creates epistemic displacement: AI doesn’t just assist with disciplinary work; it performs the epistemic labor that constitutes disciplinary thinking. The danger isn’t inaccuracy. It’s that AI can be accurate while teaching students to accept outputs generated through non-disciplinary processes.
Each discipline has core epistemic commitments:
Mathematics values logical progression and conceptual relationships
History privileges source evaluation and evidence-based interpretation
Science centers on experimental design and empirical grounding
Literary studies emphasizes close reading and attention to form
The question: Can students develop disciplinary ways of knowing when AI can perform the visible work of the discipline without engaging its underlying epistemic processes?
Territory 2: Knowledge Asymmetry and Intellectual Agency
Students with strong disciplinary grounding approach AI as a testable claim generator. Students without that grounding approach AI as an authoritative oracle.
The asymmetry compounds in both directions. Students with disciplinary knowledge use AI to accelerate learning. Each interaction strengthens their thinking because they’re actively evaluating AI’s performance. Students without that knowledge become increasingly dependent on AI to frame problems and structure arguments. Each interaction reinforces acceptance rather than developing judgment.
This creates invisible dependency. AI-assisted work can look perfectly adequate. Students complete assignments, produce acceptable outputs, receive passing grades. But they’re not developing disciplinary thinking.
The relationship is multiplicative: Disciplinary grounding × AI functional understanding = Intellectual agency.
When either factor is weak, agency collapses. This is why AI literacy must be disciplinary-specific and why it must develop alongside, not after, disciplinary knowledge.
Territory 3: Instructional Architecture
The traditional pedagogical sequence doesn’t work. We cannot teach “AI skills” first, then have students apply them. This treats AI literacy as portable, generic knowledge.
Instead, position AI as an object of disciplinary investigation from the start. Students and teachers become co-investigators of how AI performs within a domain.
Core principle: Students encounter disciplinary standards before AI outputs, then use those standards to evaluate AI’s performance.
In history, students examine primary sources first, practice sourcing, then investigate how AI handles the same material. They discover that AI synthesizes without evaluating sources, presents interpretations as facts, misses historiographical debates.
In mathematics, students work through proof construction first, then ask AI to prove the same theorem. They discover that AI can produce correct proofs that lack the conceptual connections mathematicians value.
In science, students analyze data first, practice reasoning about patterns and evidence, then investigate how AI explains the same phenomena. They discover that AI presents claims without the evidential grounding scientists require.
This creates earned skepticism. Students develop critical perspective through systematic investigation of how AI performs against disciplinary standards they’re actively learning.
Territory 4: Assessment
Traditional assessment assumes a direct relationship between what students produce and what they understand. This breaks down when AI can generate sophisticated outputs students don’t fully understand.
We must distinguish three levels:
AI-dependent work: Students outsource disciplinary thinking to AI
AI-assisted work: Students use AI as a resource within processes they direct
AI-informed work: Students investigate AI’s performance to deepen understanding
Current assessment practices can’t reliably distinguish these levels. We need new forms of evidence: process documentation, disciplinary justification, revision tasks targeting disciplinary thinking, comparative analysis of AI outputs against disciplinary standards.
The shift is from “did you complete this task correctly?” to “can you engage in disciplinary thinking?”
Territory 5: Disciplinary Authenticity and Transfer
Disciplinary authenticity isn’t about which tools professionals use. It’s about whether AI use preserves or compromises the epistemic work that defines the discipline.
Authentic use preserves the reasoning, judgment, and methodological rigor that distinguishes disciplinary expertise from information access. AI can accelerate certain tasks, provide resources, generate possibilities. But it cannot replace the evaluative frameworks and ways of knowing that constitute disciplines.
The four roles (Critic, Interlocutor, Editor, Verifier) provide consistent structure across disciplines. But what transfers is the framework, not specific practices. Students develop procedural meta-literacy: understanding that AI use requires critical evaluation, purposeful interaction, thoughtful revision, and rigorous verification, even as specifics vary by discipline.
Territory 6: Professional Development and Infrastructure
Teachers need functional understanding: practical knowledge of what AI can and cannot do within their discipline, grounded in their expertise.
A history teacher needs to know that AI synthesizes information without evaluating sources. A mathematics teacher needs to know that AI can solve problems without revealing conceptual connections. A science teacher needs to know that AI presents claims without experimental grounding.
This functional understanding develops through disciplinary investigation: teachers exploring AI’s performance on tasks central to their field, using their expertise to evaluate outputs, identifying patterns in what AI does well and poorly.
The most effective model is discipline-specific collaborative inquiry. Teachers within a discipline work together to investigate AI’s relationship to their field’s intellectual work.
Institutions need coherent frameworks: tool governance structures involving disciplinary expertise, implementation protocols that articulate principles while preserving flexibility, cross-disciplinary dialogue that helps students understand variation, administrative support that recognizes effective integration requires time and ongoing refinement.
Moving Forward
These six territories map the intellectual landscape we must navigate together. Not as problems to solve but as ongoing work that strengthens disciplinary teaching by making epistemic commitments explicit, scaffolding student investigation, and developing meta-awareness of how disciplines construct knowledge.
The work happening in Central Ohio, the commitment I witnessed at the Ohio School Boards Association conference, suggests we’re ready for this complexity. We’re ready to move beyond simplistic approaches toward frameworks that honor disciplinary thinking while preparing students for AI-saturated futures.
The question isn’t whether to integrate AI. It’s whether we’ll do so in ways that build or bypass disciplinary capabilities. These six territories give us language and structure for that essential work.
Nick Potkalitsky, Ph.D.
Check out some of our favorite Substacks:
Mike Kentz’s AI EduPathways: Insights from one of our most insightful, creative, and eloquent AI educators in the business!!!
Terry Underwood’s Learning to Read, Reading to Learn: The most penetrating investigation of the intersections between compositional theory, literacy studies, and AI on the internet!!!
Suzi’s When Life Gives You AI: A cutting-edge exploration of the intersection among computer science, neuroscience, and philosophy
Alejandro Piad Morffis’s The Computerist Journal: Unmatched investigations into coding, machine learning, computational theory, and practical AI applications
Michael Woudenberg’s Polymathic Being: Polymathic wisdom brought to you every Sunday morning with your first cup of coffee
Rob Nelson’s AI Log: Incredibly deep and insightful essay about AI’s impact on higher ed, society, and culture.
Michael Spencer’s AI Supremacy: The most comprehensive and current analysis of AI news and trends, featuring numerous intriguing guest posts
Daniel Bashir’s The Gradient Podcast: The top interviews with leading AI experts, researchers, developers, and linguists.
Daniel Nest’s Why Try AI?: The most amazing updates on AI tools and techniques
Jason Gulya’s The AI Edventure: An important exploration of cutting-edge innovations in AI-responsive curriculum and pedagogy
Recently held a faculty fete workshop on AI and one of the recommended outcomes was a discipline specific manual as a starting point. Layering on DSAIL to this manual will be an excellent next step! Thank you for sharing your work.
Just fabulous work... im days from finishing a piece that looks at this from another angle but you post is illuminating... glad there are good people thinking about this