Grade the Process of Learning as Well as the Product

September 10, 2026

Prioritize Grading the Process of Learning in the Age of Artificial Intelligence

Why Shift from Product to Process?

Generative AI can now produce essays, code, reports, and other academic products that appear highly polished. As a result, final submissions alone may no longer provide sufficient evidence of student learning.

Rather than focusing primarily on what students produce, instructors can place greater emphasis on how students learn:

  • How  do your students develop ideas?
  • How  do your students evaluate information?
  • How  do your students revise their work?
  • How  do your students respond to feedback?
  • How  do your students use AI critically and responsibly?

The goal is not to eliminate AI from learning but to assess the thinking, decision-making, and growth that occur throughout the learning process.

Key Principles

Make Learning Visible

Require your students to document important stages of their work:

  • Research notes
  • Drafts and revisions
  • Design sketches
  • Learning journals
  • AI-use statements

Reward Revision

Learning often becomes visible through improvement. Consider grading your students on how effectively they incorporate feedback and explain the changes they made between drafts.

Assess Reflection

Ask your students to explain:

  • What  did your students learn?
  • How  did their thinking change?
  • What challenges did they encounter?
  • How did AI influence their work?

Evaluate AI Literacy

Instead of asking whether your students used AI, ask how they used it:

  • What prompts did they use?
  • Which AI suggestions did they reject?
  • How did they verify information?
  • What errors did they identify with the AI’s output?

A Simple Process-Based Grading Model

A process-oriented assignment might distribute grades as follows:

  • Planning and preparation: 20%
  • Process documentation: 30%
  • Revision and improvement: 20%
  • Reflection: 15%
  • Final product: 15%

The final product still matters, but it becomes only one source of evidence.

Examples Across Disciplines

History: Research Essay

Instead of grading only the final paper, assess:

  • The development of the research question.
  • The source evaluation notes.
  • The student’s critique of AI-generated historical interpretations.
  • Their reflection on how the student’s argument evolved.

Primary evidence of learning: historical reasoning and source analysis.

Biology: Laboratory Investigation

Instead of focusing only on the lab report, assess:

  • Experimental planning
  • Observation records
  • Evaluation of AI-generated interpretations of results
  • Reflection on errors, limitations, and alternative explanations

Primary evidence of learning: scientific thinking and judgment.

Computer Science: Programming Project

Instead of grading only working code, assess:

  • Design proposals
  • Development journals
  • Documentation of AI-assisted coding
  • Oral explanation of decisions and debugging strategies

Primary evidence of learning: computational thinking and problem-solving.

Additional Assessment Types That Work Well

Learning PortfoliosStudents collect evidence of growth over time, including drafts, reflections, feedback, and revisions.
Process JournalsStudents maintain a running record of decisions, challenges, breakthroughs, and AI use.

Oral Defenses		Students explain and defend their work through short presentations, interviews, or project demonstrations.
These approaches make it easier to assess understanding even when AI was involved in producing the final work.

Try This:

Use the following prompt with your preferred AI tool to generate process-based versions of existing assignments:

“I teach [COURSE/DISCIPLINE].

Here is my current assignment:

[PASTE ASSIGNMENT]

Redesign this assignment so that at least 70% of the grade is based on evidence of the learning process rather than the final product.

Include:

– 3-5 developmental checkpoints

– Student reflection activities

– Opportunities for revision

– Appropriate use of AI by students

– A rubric that emphasizes thinking, decision-making, and growth

– Suggestions for keeping instructor workload manageable

Explain why each component provides evidence of learning that is difficult to outsource to AI.”

AI challenges educators to reconsider what grades are intended to measure. By assessing planning, reflection, revision, decision-making, and responsible AI use, instructors can focus on what ultimately matters most: evidence of student learning and intellectual growth.

←back
Next: Develop Rubrics with AI Assistance