Learning approach

Learners decide. AI builds. Real people judge.

One framework runs every Arena School program, for students, teachers, school leaders, and parents. Here is what learners practice, what they can do by the end, and how we know.

Framework

One framework, every program.

The learner sits at the center. Four strands of work circle them. Four teaching principles hold it in place.

Four strands
  1. Inquiry and planningTurn an interest into a focused question, and plan the work.
  2. Research and evaluationCheck sources, including AI output, and choose on evidence.
  3. Creating and communicatingBuild and test it with AI, and explain it clearly.
  4. Collaboration and reflectionUse critique from teachers and peers to revise the work.
Four teaching principles
  1. In person, real-time feedback
  2. AI use disclosed and cited
  3. Learner voice and choice
  4. An authentic audience
The teaching principles in full →
The question at every check-in

How did you check it?

We say learner on purpose: the same framework runs our student programs, our teacher training, and our sessions for school leaders and parents. Built on project-based learning and informed by PBLWorks’ Gold Standard PBL. AI can support the feedback. It does not replace it.

One week

Five days, all four strands.

A five-day student build week, as run at the International School of Beijing in summer 2026. Darker blocks show the day each strand is practiced.

StrandDay 1Pick a problemDay 2Go publicDay 3The rebuildDay 4Polish and priceDay 5Demo dayEvidence
01Inquiry and planningTurn an interest into a focused questionDay 1
Day 1Pick a problem and a plan
EvidenceA plan and a first prototype
02Research and evaluationCheck sources, including AI outputDays 1 to 3
Day 1In progressDay 2Test it with the people it’s forDay 3In progress
EvidenceChoices backed by what users did
03Creating and communicatingBuild and test a product with AIDays 2 to 4
Day 2Ship itDay 3In progressDay 4Price it
EvidenceA live product at a public link
04Collaboration and reflectionUse critique to revise the workDays 3 to 5
Day 3Rebuild itDay 4In progressDay 5Defend it
EvidenceA rebuilt product, and a pitch to judges

No worksheet, no quiz. The work is the evidence.See the five-day Entrepreneurship & AI Intensive →

What learners will be able to do

Outcomes you can assess on any project.

The same outcomes hold for every learner. The depth grows, from middle school to adults.

Inquiry and planning

  • Turn a broad interest into a focused, researchable question.
  • Plan a project with milestones, and adjust the plan based on progress.

Research and evaluation

  • Find, evaluate, and cite sources, including checking the accuracy of AI output.
  • Compare options against criteria, and justify a choice.
  • Explain how AI tools work, where they fail, and the bias and privacy issues they raise.

Creating and communicating

  • Build and test a product, prototype, or solution, deciding which steps to hand to AI and disclosing how they used it.
  • Structure an argument and write it clearly, in their own words.
  • Present to an authentic audience, and defend their reasoning under questioning.

Collaboration and reflection

  • Work in a team, share roles, and contribute to a shared product.
  • Give and use critique from teachers and peers to revise their work.
  • Reflect on their process and on what they would do differently.
Assessment

We assess the thinking behind the work.

The process, not only the product: how a learner chose, checked, and finished.

Sample criterion · Research and evaluation

Evaluates sources, including AI output

  1. BeginningAccepts sources and AI output as given; cites inconsistently.
  2. DevelopingChecks some claims against a second source; cites most sources.
  3. ProficientVerifies key claims, including AI output, against reliable sources; discloses AI use.
  4. ExtendingExplains why sources disagree, corrects AI errors, and judges which evidence to trust.

Used with students in the ISB AI Founders Program, summer 2026. With a school, the rubrics stay with its team.

Evidence we collect
  • Process journal
  • Drafts
  • Recorded defense
Three questions do the coaching
  • AgencyWhy this problem?
  • TasteHow do you know it’s good?
  • CraftWhat would make someone trust it?
Why we teach this way

The skills and principles behind the framework.

Our approach explains why agency, taste, and craft come first, the principles in every room, and how the framework fits students, teachers, schools, and families.

  • Three skills
  • How we teach
  • Who does what
  • By audience
Read our approach →
Next step

See it in a program.

From two-day workshops to year-long mentorship, every Arena School program runs on this framework.

Judges test a student product before scoring
Judges test each product before scoring, Beijing 2026