AI for EverybodyLearn the intelligence era with better judgment.

Your progress: 0 complete

Glossary My Map AccountQuick answers Choose my route History of AI Sources About this platform

You are here

Current learning moveUnderstand

Follow material in an order that builds understanding.

Do nowStart this unit’s learning plan

Your guide

Assessment after generative AI

Your learning move

UnderstandBuild ideas in a deliberate order.

Why this page exists

Learn is for subjects where sequence matters. The foundation comes first, the core adds depth, and thematic paths apply the ideas to a direction.

What you will leave with

A polished product is weaker evidence when assistance is easy to obtain and hard to observe. Assessment can respond by collecting several kinds of evidence across process, explanation, application, sources, decisions, and revision.

Before you begin

Every path states its level and time. AI from Zero assumes nothing.

Do this now

Choose the earliest route that feels new to you; do not start with the longest course by default.

Do now on this pageStart this unit’s learning planThe plan states the outcome, time, prerequisite, immediate action, and prepared continuation.Another valid routeCompare every pathChoose by outcome, time, and subject rather than card position.

Account

Education, Creativity, and Human Capability

Unit 03 · evidence of learning

Assessment after generative AI

Plan for this page

Understand the idea, test it actively, then explain it in your own words.

You will leave with: compare process, oral, project, and evidence-based assessment align permitted AI assistance with the capability being assessed design a proportionate set of learning evidence without surveillance by default
Time: 30 min
Before you begin: General AI Literacy is useful preparation, but not required. The path is designed for students, educators, parents, creators, and anyone learning with AI.
Do this now: Read the outcomes and useful terms before the first section.

Prepared next stepCreativity and authorship

After this unit, you should be able to

compare process, oral, project, and evidence-based assessment
align permitted AI assistance with the capability being assessed
design a proportionate set of learning evidence without surveillance by default

Terms you will use

These definitions prepare you for the reading; you do not need to memorize them.

Construct

The knowledge, skill, judgment, or capability an assessment is intended to reveal.

Process evidence

Visible traces of planning, attempts, decisions, sources, feedback, revision, and explanation produced while work develops.

Assessment validity

The degree to which the available evidence supports the intended interpretation about learning, within a stated context and set of limitations.

Begin with the claim about learning

An assessment should first state what it is trying to show. If the goal is independent recall, access to generated answers changes the construct. If the goal is critical editing, responsible tool use may be part of the construct. If the goal is collaboration, a solitary timed response may miss important capability.

The same AI rule cannot serve every purpose. An educator can specify which stages must be independent, which tools are permitted, what disclosure is expected, and what evidence will distinguish learner understanding from a system-produced artifact.

Combine complementary evidence

Process evidence can include a question proposal, plan, source log, early attempt, decision note, feedback response, version comparison, or reflection on a failed approach. Oral evidence can ask the learner to explain a choice, apply an idea to a new case, or respond to a challenge rather than merely recite a prepared answer.

Projects can reveal sustained inquiry, construction, collaboration, and revision when milestones and responsibilities are visible. Evidence-based assessment can require claims to be linked to sources and limitations. No format is automatically valid: each needs accessible instructions, appropriate conditions, and a clear link to the intended capability.

Prefer transparency to unexamined detection

An automated label is an inference rather than direct observation of how a submission was produced. Its usefulness depends on evidence about the particular tool, population, language, assistance and conditions. This path has reviewed no detector for disciplinary proof, so a score cannot replace the institution’s fair procedure and relevant evidence.

A more educational response uses clear expectations, proportionate documentation, conversation, and multiple observations over time. Privacy matters: collecting keystrokes, screens, drafts, or biometric data creates its own risks. Gather only evidence needed for the learning claim and explain how it will be used and retained.

Interactive model

Four windows onto learning

Selected layerProcess

Plans, attempts, source choices, decision notes, feedback, revisions, and division of work.

Read the complete text alternative

ProcessPlans, attempts, source choices, decision notes, feedback, revisions, and division of work.
OralExplanation, questioning, reconstruction, comparison, and application to an unprepared case.
ProjectSustained creation or inquiry with milestones, constraints, collaboration, and authentic consequences.
EvidenceTraceable claims, appropriate sources, uncertainty, limitations, and reasons for accepting or rejecting material.

Misconception to correct

“The solution to generative AI is one perfect detector that proves who wrote every sentence.”

This path has reviewed no evidence for such a detector. Clear rules, fair procedure, and several direct forms of learning evidence provide a stronger assessment design than an unexamined score.

Active checks

Decide, then compare the reasoning

No grade, score, or streak. Progress records completed learning objects, not your worth or ability.

A course aims to assess whether students can critique evidence and revise an argument. Which design is best aligned?

An automated detector labels a multilingual student’s work as likely AI-generated. What should happen next?

Explain back and transfer

Redesign one assessment

Choose an existing task. State the capability, allowed and prohibited assistance, then combine at least two evidence windows. Add an accessible alternative, disclosure rule, privacy limit, and fair response to uncertainty.

Answer both checks before completing.

Source context

What supports this unit and where it stops

Evidence status: emerging

AI Competency Framework for TeachersUNESCO

Provides a teacher-competency framework linking AI foundations, ethics, pedagogy and professional learning; it does not mandate one assessment policy across subjects or education systems.

Open source

AI Competency Framework for StudentsUNESCO

Emphasizes human-centred judgment, ethics, application and creation; local age, curriculum, language, disability and access contexts still shape valid assessment.

Open source

Day of AIDay of AI

A curriculum and educator-resource platform demonstrating hands-on activities, reflection and stakeholder discussion; individual materials require review and do not establish universal assessment effectiveness.

Open source

Loading page content...

…