LENS Learning Engineering for Next-Generation Systems
Prototypes

Problems → Prototypes · AILE

← Back to this problem brief

A Context-Rich AI Language Practice Companion

Language apps can make practice frequent and convenient, yet generic drills may not prepare immigrant and multilingual learners for the specific academic, social, and institutional situations where language must be used.

Try it

This exercise runs locally using simple rules. Your entries stay on this page and clear when you reload. No AI service is called.

Local practice controls

Practice a classroom conversation

Choose an authored English classroom task or let a teacher select one. Add your own context and plan in any language. Support appears after your first attempt. These are fixed hints and role-play turns; no AI, translation, speech recognition, pronunciation feedback, or proficiency scoring is provided.

Your situation

Your teacher says: “Compare the two characters and support your answer.” You are unsure what kind of support to include. Ask a question that makes the unclear part specific.

Type a response or a transcript of an oral rehearsal. No recording is made. Changing the task, context, notes, or first attempt clears the revision stage.

Use invented or non-identifying details. Nothing is saved automatically. Downloads contain your text; you decide whether to share them.

Critical review of this working prototype

Strongest assumption: Rehearsing specific communication functions and revising after an attempt transfers to real classroom participation.

Likely failure: Learners copy an English model, mistake authored support for personalized evaluation, or reuse a revision after switching to a different task.

Test that could change the design: Have a willing teacher observe an unfamiliar instruction or group task after practice. Reject transfer claims if rehearsed text improves but listeners still cannot understand the learner's intended question or contribution.

Non-AI comparison: Teacher-selected role-play cards with a peer, home-language planning, and a communication checklist.

Evidence to collect: Human-observed communicative success on an unseen task, repair questions, participation, and learner confidence compared with observed understanding. No unvalidated proficiency or grammar score.

Human control and access: English authored models cannot validate other languages, pronunciation, or fluency. Text and oral-rehearsal transcripts plus home-language notes are accepted; there is no audio capture. All review is voluntary and records stay local until downloaded.

Changes made during review

  • Added three distinct task situations, hints, function checklists, and authored partner turns.
  • Gated support on a nonempty first attempt and labeled it fixed authored support.
  • Required a revision, partner-turn reply, self-check, and real-world transfer plan.
  • Cleared support and revision stage when the initial task, context, home notes, or attempt changes.
  • Preserved original and revised responses for human comparison without proficiency scoring.
  • Disabled no-JavaScript input/submission and completed all eight design-cycle steps.
  • Smallest next experiment: peer observation on an unfamiliar classroom instruction, followed by learner reflection.

This is a design and implementation review, not an empirical validation of learning outcomes.

Optional AI critique

Take the brief into your own AI environment

Nothing is sent until you choose.

Copy a self-contained review prompt, then open the AI workspace you already use. The prompt asks for a falsification test, a non-AI alternative, evidence that distinguishes activity from learning, and human-system risks.

Open ChatGPT ↗ Open Claude ↗ Open Gemini ↗

Preview the prompt 
Act as a critical learning-engineering reviewer. Review the prototype brief below as a thin-slice experiment, not as a finished product.

Return:
1. The strongest design assumption.
2. The most important plausible failure mode.
3. The highest-value falsification test for the next cycle.
4. One credible non-AI alternative that could address the same capability gap.
5. The next smallest prototype worth building.
6. What should be measured to distinguish engagement from learning, transfer, and real performance.
7. Human-agency, equity, accessibility, privacy, or governance concerns that should change the design.

Be concrete. Separate evidence-backed claims from hypotheses. Do not reward novelty for its own sake, and do not assume more AI is better.

COLLECTION: AILE
TITLE: A Context-Rich AI Language Practice Companion

CAPABILITY GAP:
Language apps can make practice frequent and convenient, yet generic drills may not prepare immigrant and multilingual learners for the specific academic, social, and institutional situations where language must be used.

FIRST PROTOTYPE:
Create a teacher-configurable AI language companion grounded in real school and community tasks. It supports speech, text, images, home-language clarification, and repeated practice while keeping the target communicative function explicit.

REPRESENTATIVE USE CASE:
A newcomer practices asking a teacher for clarification, interpreting an assignment instruction, and contributing one idea to a group discussion before trying the same tasks in class.

LEARNING / HUMAN-SYSTEM FRAME:
The design connects language learning to authentic use, builds on the learner's existing linguistic resources, and offers multiple representations and response modes. AI provides practice volume; educators determine goals and interpret progress.

QUESTIONS ALREADY IDENTIFIED:
1. Which simulated situations transfer most strongly to real participation?
2. When is home-language support helpful versus overused?
3. Does practice increase confidence and actual classroom communication?

FIRST-CYCLE FRAMING:
Understand: Validate that the real gap is the ability to use a second language successfully in authentic academic and social situations, not simply low engagement, low tool use, or a workflow inconvenience.
Map: Map the system: learner, teacher, peers, home language, classroom/community tasks, AI language companion, and school norms. Identify where the capability currently succeeds, breaks down, or is masked by other constraints.
Instrument: Predefine evidence: communicative success, vocabulary/structure growth, classroom participation, transfer, confidence calibration, and language-support use. A key disconfirming signal is: app performance improves without more successful communication in real settings.
Open the underlying design brief and eight-step path

A first prototype

Create a teacher-configurable AI language companion grounded in real school and community tasks. It supports speech, text, images, home-language clarification, and repeated practice while keeping the target communicative function explicit.

Representative use case

A newcomer practices asking a teacher for clarification, interpreting an assignment instruction, and contributing one idea to a group discussion before trying the same tasks in class.

Questions worth carrying forward

  • Which simulated situations transfer most strongly to real participation?
  • When is home-language support helpful versus overused?
  • Does practice increase confidence and actual classroom communication?

Evidence anchors

These sources motivate the design; they do not validate this prototype.

LENS iteration cycle

One possible first pass

Refine → Understand
  1. 01
    Understand

    Validate that the real gap is the ability to use a second language successfully in authentic academic and social situations, not simply low engagement, low tool use, or a workflow inconvenience.

  2. 02
    Map

  3. 03
    Design

    Compare an AI intervention with simpler non-AI options. Specify what the human decides, what the AI may suggest, and which tradeoffs are acceptable.

  4. 04
    Build

    Create the smallest usable prototype around one representative task, with expert-curated content, visible uncertainty, and an easy human override.

  5. 05
    Instrument

    Predefine evidence: communicative success, vocabulary/structure growth, classroom participation, transfer, confidence calibration, and language-support use. A key disconfirming signal is: app performance improves without more successful communication in real settings.

  6. 06
    Deploy

    Pilot with a small, representative group in a near-real setting; preserve a baseline or comparison condition and document implementation conditions.

  7. 07
    Evaluate

    Look for capability growth and transfer, not just satisfaction or activity. Inspect subgroup patterns, human workload, errors, and unintended adaptations.

  8. 08
    Refine

    Let the evidence change the problem model. Default back to UNDERSTAND if the assumed gap was wrong; otherwise revisit the earliest step invalidated by the evidence.