In 1949, educator Ralph Tyler posed two questions that still resonate with educators today: What do I want students to learn? What evidence would I accept to verify their learning?
Tyler’s questions framed much of 20th century curriculum design: define clear objectives, then assess whether students meet them. For decades, this model worked well in a world where content mastery, recall, and procedural fluency were paramount.
But today, the answers to Tyler’s second question (what evidence we accept as learning) are less clear. In an AI-enhanced world, students can generate full essays, solve math problems, and produce visuals with generative tools. This means that what we assess, how we assess, and what we value as evidence of learning must evolve. In this new environment, it’s more important than ever for students to develop skills that are not just about content, but about thinking, questioning, interpreting, and adapting.
Essential AI Skill Set
These 10 skills can help students thrive in a world increasingly shaped by generative AI:
- Information literacy: Finding, evaluating, and using information effectively and responsibly. This skill helps students distinguish credible content from AI-generated noise and misinformation.
- Data literacy: Interpreting, analyzing, and communicating with data in meaningful ways. Students with this skill question AI-generated statistics, identify misleading trends, and make informed decisions.
- Questioning: Asking purposeful, focused, and open or closed questions to drive inquiry or direct an AI tool. This skill empowers students to engage deeply with AI and guide their own learning.
- Prompt engineering: Crafting clear, precise inputs to generate useful, accurate AI responses. Strong prompts help students communicate effectively with AI systems to get results aligned to their goals.
- Dialogue: Exchanging ideas back and forth with AI to build meaning through clarification and reasoning. Students use this skill to improve AI interactions and deepen their understanding.
- Verification: Checking the accuracy, reliability, and source of AI-generated information. This skill cultivates critical thinking and helps students detect bias or misinformation.
- Critical interpretation: Analyzing AI outputs for logic, clarity, and relevance. This teaches students to go beyond surface-level acceptance and assess the reasoning of what they’re reading or generating.
- Curiosity: Being driven to explore uncertainties and pursue new knowledge. This intrinsic motivation ensures students use AI as a tool to expand their thinking, not shortcut the learning process.
- Metacognition: Reflecting on one’s own thinking and adjusting strategies as needed. Metacognitive learners evaluate when and how to use AI effectively, rather than relying on it passively.
- Cognitive flexibility: Adapting one’s thinking, shifting perspectives, and approaching problems from multiple angles. This skill supports creative and resilient AI use when situations, inputs, or goals change.
What These Skills Mean for Teaching
These AI skills aren’t supplemental to the core standards students must learn. They’re now foundational to the learning students must do. To develop them, educators need to shift what they teach and how they assess. Traditional essays, quizzes, and projects have value, but they must be layered with tasks that involve prompting AI, critiquing its output, and reflecting on the learning process.
Instructional design must now include how students use AI, intentionally and responsibly, as part of their cognitive toolkit. Students should be encouraged to explain their process: How did they use AI to start their thinking? What parts of the AI-generated response did they accept, reject, or revise? What additional verification steps did they take?
Rubrics and success criteria must also adapt. Instead of only measuring final products, teachers might assess the quality of students’ prompts, the logic of their revisions, or their reasoning during peer discussions.
In the video accompanying this column, high school CTE business teacher Sarah Hawley models this shift. At Health Sciences High, every assignment and assessment includes an AI usage code from 0 to 3, adapted from the North Carolina Department of Public Instruction:
- AI Free: Students must complete all work independently, with no AI assistance permitted.
- AI Assisted: AI may be used for planning or brainstorming, but final submissions must be AI-free.
- AI Enhanced: Students use AI interactively to support learning, with required human oversight and critical engagement.
- AI Empowered: AI is fully integrated to enable creativity, problem solving, and innovation with students responsible for quality and originality.
This system teaches students to make intentional choices about when and how to use AI, and it guides teachers in shaping expectations accordingly. In the personal finance lesson from the video, students are each assigned a persona who has budget needs. They dialogue with AI systems and consider various options. Students must critique the AI bot’s suggestions, identify gaps, revise outputs using real-world data, and explain how the tool helped or hindered their process. The AI code for this task is set to 3, meaning students are expected to engage directly with AI as part of the learning experience.
Why AI Skills Matter
Educators today must lead with purpose, designing learning experiences that prepare students to question and shape AI. We no longer live in a world where having the right answer is enough. In the age of AI, students need the right skills to ask, to analyze, to adapt, and to assess—and teachers can develop the AI skills necessary to foster this human intelligence.
Instructional Insights / November 2025 / Exploring the Ethics of AI with Students
in 1 week
Video Reflection: AI in a Personal Finance Lesson
After watching the video, consider the following questions for reflection or discussion with your colleagues.
- What AI skills are students developing and practicing?
- How does Ms. Hawley’s lesson differ from a pre-AI version of this task, and what aspects remain the same?
- What features of her task design make it more AI-resistant?
