Purpose
Is the framework a competency model, developmental continuum, pedagogical model, policy framework, assessment model, critical-theory framework, or institutional planning tool?
Research • Teaching • Practice
AI literacy should teach us to question the machine, not merely operate it.
Critical AI Literacy is a growing resource hub for educators, students, librarians, and researchers examining how artificial intelligence shapes knowledge, learning, power, and decision-making.
Start Here
Critical AI literacy goes beyond knowing how to use the tool.
This section introduces the ideas that shape the rest of the site. If you are new to critical AI literacy, begin here before exploring individual frameworks, research studies, or teaching resources.
What Is Critical AI Literacy?
AI literacy is often described as the knowledge and abilities people need to understand, use, evaluate, and communicate about artificial intelligence. Those abilities matter, but they do not tell the whole story.
Critical AI literacy asks additional questions.
It asks learners not only whether an AI system can perform a task, but also how the system produces its answers, what evidence supports those answers, whose knowledge is represented, what assumptions are embedded in the technology, who benefits from its use, and what consequences may follow.
A critically AI-literate person should be able to move among several kinds of thinking
- Understanding: Recognize what AI systems can and cannot do and develop a realistic understanding of how generative AI produces output.
- Evaluation: Assess claims, citations, reasoning, evidence, uncertainty, and reliability rather than treating fluent output as trustworthy by default.
- Context: Consider whether AI use is appropriate for a particular discipline, assignment, audience, research question, or professional setting.
- Critique: Examine issues such as bias, power, equity, privacy, surveillance, labor, accessibility, environmental impact, and epistemic authority.
- Transparency: Communicate meaningful information about how AI contributed to a piece of work.
- Agency: Make deliberate decisions about when to use AI, how much responsibility to delegate to it, and when another approach is preferable.
Critical AI literacy therefore involves both competence and judgment. Someone may be highly skilled at prompting an AI system and still have limited understanding of its limitations, evidence base, social consequences, or appropriate use.
It is also not a fixed checklist that someone completes once. Critical AI literacy develops through practice and changes with context. The knowledge and judgment needed by a first-year college student may differ from those needed by a faculty researcher, librarian, doctoral student, healthcare professional, or institutional leader.
The goal is not to produce either enthusiastic AI adopters or automatic AI rejecters. The goal is to help people make informed, evidence-based, transparent, and responsible choices about artificial intelligence.
Why frameworks differ
There is no single universally accepted AI literacy framework.
Different frameworks are designed to solve different problems. Some identify the competencies people need. Others describe how AI literacy should be taught, how development should be assessed, how institutions should create policy, or how learners should critically examine the social and political systems surrounding artificial intelligence.
As a result, two frameworks may both use the term AI literacy while emphasizing very different things. One may focus heavily on understanding AI concepts, using tools, writing prompts, and evaluating output. Another may devote much more attention to bias, power, labor, privacy, equity, environmental consequences, or who has authority to produce knowledge.
Population
Is it designed for students, faculty, librarians, researchers, staff, professionals, institutional leaders, or multiple populations?
Critical Depth
How deeply does it address bias, power, equity, labor, privacy, surveillance, accessibility, environmental impact, epistemic authority, and human agency?
Evidence Base
Is it theoretical, empirically derived, review-based, validated through research, policy-based, or developed through expert consensus?
Portability
How well might it travel across community colleges, research universities, doctoral programs, disciplines, and national contexts?
Usefulness
Can it support curriculum, faculty development, assignment design, policy, library instruction, research design, coding, or assessment?
A Note About the Term “Critical”
On this site, critical does not simply mean being skeptical of AI or looking for errors. Critical AI literacy includes verification and skepticism, but it goes further. It asks how artificial intelligence participates in larger systems of knowledge, authority, technology, economics, education, and power.
That means asking questions such as who designed the system, what data shaped it, whose knowledge is represented or excluded, what kinds of labor make the system possible, who has access to the technology, what happens to personal or institutional data, what environmental resources the system consumes, how AI may change whose expertise is trusted, and who is accountable when an AI-supported decision causes harm.
Critical AI literacy therefore combines technical understanding, information evaluation, ethical reasoning, social analysis, and human judgment. That broader view provides the foundation for the frameworks, scholarship, teaching resources, and research collected throughout this site.
A working model
Six dimensions of critical AI literacy
Understand
Recognize how generative AI systems work, what they produce, and where their limitations begin.
Interrogate
Question assumptions, training data, design choices, incentives, and claims of neutrality.
Verify
Check claims against reliable evidence and trace information back to real, appropriate sources.
Contextualize
Consider disciplinary norms, audience, task, institutional policy, and consequences of use.
Disclose
Communicate AI use transparently enough for others to understand its role in the work.
Decide
Make deliberate choices about when AI helps, when it harms, and when another approach is better.
Frameworks & scholarship
Map the field, not just the tools.
This section will collect and compare critical AI literacy frameworks, conceptual models, empirical studies, and emerging scholarship across higher education.
Coming first
Framework comparison
A crosswalk of major AI literacy and critical AI literacy frameworks, including populations, critical depth, evidence base, and practical uses.
Follow the project →In development
Critical dimensions index
A structured way to examine whether frameworks meaningfully address bias, power, privacy, surveillance, labor, equity, environmental impact, and epistemic authority.
See what is being built →Research library
Selected scholarship
A curated and annotated collection of research for instructors, librarians, students, and researchers who need more than an undifferentiated bibliography.
Explore the research plan →Teaching & learning
Resources designed for actual classrooms.
The teaching collection will emphasize decisions, verification, transparency, and transfer rather than memorizing a parade of rapidly changing tools.
AI Use Decision Tree
Help learners decide whether, when, and how AI belongs in a particular task.
Planned resourcePrompt-to-Verification Ladder
Move from generating an answer to testing the claims, evidence, citations, and omissions behind it.
Planned resourceTransparency Statement Builder
Turn vague “I used ChatGPT” disclosures into meaningful descriptions of what AI contributed.
Planned resourceAssignment Retrofit Checklist
Redesign assignments around learning evidence, process, judgment, and responsible AI choices.
Planned resourceVerification
Trust is not a research method.
Generative AI can produce fluent language without providing dependable evidence. Verification is therefore a core literacy practice, not an optional final check.
Extract the claim
Identify what the AI output is actually asking you to believe.
Locate independent evidence
Move outside the generated response. Search scholarly databases, authoritative sources, primary materials, or other evidence appropriate to the claim.
Verify the citation
Confirm that sources exist, say what the response claims they say, and are appropriate for the question.
Look for what is missing
Check whose perspectives, counterevidence, uncertainty, context, or limitations have disappeared.
Reassess the conclusion
Revise your judgment based on evidence rather than allowing the AI's confidence or fluency to substitute for it.
Research
A living research hub.
The site is designed to grow alongside ongoing work on critical AI literacy in higher education, including research across institutional types and learner populations.
Research focus
What support do people need to develop critical AI literacy?
Students
Learning, research workflows, verification practices, transparency, confidence, and responsible use.
Faculty & instructors
Teaching needs, assignment design, classroom expectations, policy, and support for student development.
Institutions
How approaches travel across community colleges, research universities, doctoral programs, and different national contexts.
About the project
Built to remain useful after today's AI tool is obsolete.
CriticalAILiteracy.net is an independent educational and research resource focused on critical approaches to artificial intelligence literacy in higher education.
The site will bring together research synthesis, practical teaching materials, verification tools, framework analysis, and resources that can be adapted across institutions. Its goal is not to prescribe one approved relationship with AI. It is to make better-informed choices possible.
This site is currently in its first build. Resources, citations, and project documentation will be added as the collection develops.
Critical AI literacy