From Structuring to Alignment: Clarifying Before Generating

Position in Module#

• Unit: 2
• Topic: From Structuring to Alignment: Clarifying Before Generating
• Literacy perspective: Interaction, secondarily Human-Context
• Collaboration level: Alignment
• Fundamentals focus: Generative AI Fundamentals
• Why this Friday matters: This Friday develops the unit's collaboration move through the specified literacy lens and prepares participants for the associated evidence-based challenge.

Literacy Perspective#

• Primary literacy: Interaction
• Secondary literacy, if any: Human-Context
• What this literacy explains: It explains the main dimension through which participants can predict, influence, and interpret AI behaviour in this unit.
• What participants should manipulate, observe, and interpret: Participants manipulate the unit's experiment focus, observe the stated outcomes, and interpret results through the literacy perspective.

Collaboration Level#

• Level relation: Alignment
• Collaboration move being learned: Participants practice the unit's collaboration move as a transferable way of working with AI.
• What participants should be able to do differently after this Friday: They should be able to design a more controlled AI-supported interaction, process, decision arrangement, or workflow in their own context.

Generative AI Fundamentals#

This Friday should deepen the Unit 1 introduction by focusing on context as the condition under which the model predicts. The goal is for participants to understand why clarification changes behaviour before any output is produced.

• Explain that the model predicts from everything it is given, not only from the last sentence typed by the user. The current input state can include the conversation so far, task framing, examples, constraints, assumptions, planning artefacts, and any other context inserted by the system or user.
• Use this to explain alignment-first interaction: clarification changes the future output because it changes the input state from which the model continues. A direct-start prompt and an alignment-first prompt are not merely different social behaviours; they create different predictive conditions.
• Show how context resolves ambiguity. For example, an ambiguous term such as Bank is interpreted differently in a context of sitting and public space than in a context of finance and interest rates. The same principle applies to professional tasks: audience, goal, constraints, and success criteria steer interpretation.
• Use the language-play article as a source for context-based meaning reconstruction, but leave detailed tokenisation for Unit 3. The relevant Unit 2 point is that meaning is inferred from neighbouring linguistic and task context, not from isolated words alone.
• Connect this to planning artefacts: a clarified brief, requirement list, or success-criteria document is not just documentation for humans. When passed back into the interaction, it becomes context that shapes the model's next predictions.
• Caveat to demonstrate: context is limited and unevenly influential. As a session grows, earlier instructions, details, or constraints may lose practical weight or be displaced by newer context, so participants should not assume that something said once will reliably govern all later outputs.
• Emphasize the central learning point for Unit 2: alignment works because it deliberately improves the context before generation, making the model's continuation more likely to fit the actual task, audience, and constraints.

Recurring Agentic Coding Demonstration#

Demonstrate clarification-first development by having the coding agent identify missing requirements, assumptions, edge cases, non-goals, and acceptance criteria before implementation. Turn the exchange into a reusable markdown specification or issue brief that defines scope, constraints, test expectations, and success conditions, then compare this aligned version with a direct-start attempt on the same task. Inspect whether the clarified specification improves task fit, reduces ambiguity, and produces a more reliable starting point for later coding sessions.

Friday Activity Table#

Experiment Focus#

• Experiment focus: clarification before action
• Manipulated variable, condition, or design decision: Use the Friday activity table and challenge focus to define the comparison.
• What remains constant across comparison: Keep the underlying task or use case stable enough that observed differences can be interpreted credibly.

Observation Focus#

• Observation focus: improved shared understanding and task fit
• Evidence participants should capture: prompts, outputs, traces, intermediate artefacts, screenshots, feedback, test results, or workflow records as appropriate for the unit.
• Interpretation participants should be prepared to make: Explain how the manipulated condition changed the observed outcome and what this reveals about control.

Challenge Kickoff Deliverables#

Participants must leave Friday with:

• selected context-relevant task/use case
• baseline definition for comparison
• target condition or workflow idea
• first protocol draft covering objective/use case and intended output, manipulated design choices or variables, what remains constant across comparison, expected effects, and evaluation criteria
• first captured evidence, such as one run, one workflow sketch, one retrieval trace, one judge report, one responsibility protocol, or one capability map

The kickoff should already reflect the challenge-focus logic:

• Difficulty: Direct generation fails when assumptions, constraints, audience needs, and success criteria remain implicit.
• Approach: Compare a direct-start baseline with an alignment-first variant, make clarification explicit before production, and build a reusable planning artefact that captures objective, audience, assumptions, constraints, success criteria, and control checks.
• Artefacts: protocol document; baseline interaction and output; alignment interaction evidence; planning artefact from the alignment phase; output produced from the aligned plan; comparative run log; concise conclusion on when clarification is worth the added step.

Teacher Setup#

Mandatory:

• one or more examples that make the core distinction visible
• demonstration setup for the recurring agentic coding case, if used
• runnable environment(s) with low setup friction, plus fallback option
• protocol template and evidence-capture template

Configurable:

• exact tools/platforms
• work mode: individual, pair, or group
• scaffolding depth and peer-feedback format
• whether participants implement a workflow or tightly specify it when implementation is not feasible