AIYING | V2.0 — GENERATIVE VIEW
What If the Workshop Began When It Ended?
By Ravindra Dastikop
An AI workshop ends. Participants leave with prompts, examples, and enthusiasm.
In our illustrative situation, one participant applies the learning to a real task. Another encounters a difficulty and stops. Others have yet to find an opportunity.
V1.0 examined the distance between attending a workshop and applying its learning.
Now we ask:
“Where could a situation acquire the capacity to generate its own next possibilities?”
Applied here: Where could this workshop acquire the capacity to generate its own next learning opportunities?
The hidden assumption
A familiar workshop design assumes that the trainer determines the content, participants experience it, and follow-up helps them apply what was taught.
That is a useful starting arrangement. But it leaves something underused: what happens when participants attempt their own tasks.
Their adaptations, mistakes, checks, and unanswered questions could become material for further learning.
PO: The workshop begins when it ends.
Taken literally, this is a contradiction. We use it as a deliberate provocation.
The movement is to shift the starting point of the next learning cycle.
Instead of beginning again with another prepared topic, begin with what participants discover through application.
A participant’s difficulty becomes a question to investigate. A successful adaptation becomes a worked example. An unexpected result becomes something to test.
The switch: participants’ attempts become inputs to the next round of learning.
A workshop that produces its next brief
Imagine that each participant leaves with one relevant task to attempt.
Within seven days, they return a short learning note:
- I attempted: the task and its intended purpose.
- I produced: an output or an account of what happened.
- I checked: how I assessed the result.
- I still need: one unresolved question.
These notes give the group something concrete to work with.
Suppose a teacher uses AI to draft a quiz but finds an ambiguous question. Another participant produces a useful summary but cannot verify a claim. A third adapts an exercise successfully and explains the changes.
These are hypothetical examples. Each could supply a different next activity: improving question quality, checking claims, or testing whether an adaptation works elsewhere.
The next session gains its brief from actual attempts.
What gives the situation this capacity?
The proposed generative mechanism is a recurring feedback cycle:
Attempt → evidence → shared question → next experiment.
Several parts matter.
The attempt connects learning to a real need. Evidence makes discussion inspectable. Sharing allows another person to recognise a relevant problem. The next experiment turns discussion into further action.
A collection of uploads alone would not establish this capacity. The important step is whether something contributed by a participant helps shape and improve a subsequent attempt.
People remain central: they choose worthwhile tasks, judge evidence, and decide what deserves further attention.
Where AI could help
With participants’ permission and suitable handling of their material, AI could help organise learning notes, suggest related questions, or draft a practice activity from a recurring difficulty.
A reusable prompt might be:
“Review these anonymised learning notes. Separate reported observations from interpretations. Identify recurring difficulties and suggest three small follow-up experiments. For each, specify the task, evidence to collect, and a way to check the result. Do not invent outcomes or treat repeated claims as verified facts.”
A facilitator would review these suggestions against the original notes.
AI’s contribution would be to help the group work with its experience. Participants’ evidence and human judgment would determine what happens next.
Start with one small test
Try this with a voluntary group of five participants over two weeks.
Week one: Each person attempts one task and submits the four-part learning note.
Week two: The group selects one question from those notes, designs a short follow-up exercise, and tries it.
Look for observable evidence:
- Did a participant’s attempt reveal a useful next question?
- Did that question change the follow-up activity?
- Could someone use another participant’s contribution?
- Did the second attempt show an improvement against an agreed criterion?
For example, if the problem is ambiguous quiz questions, the criterion might be whether an independent reviewer can identify one defensible answer and explain why the alternatives are incorrect.
If few people return, tasks prove unsuitable, or the notes do not help anyone act, revise the arrangement. Those results would reveal limits in the proposal.
What could emerge?
If the cycle proves useful, the group could gradually build a reviewed collection of cases: tasks attempted, approaches tried, checks performed, and questions still open.
That collection could support new participants, reveal gaps in the original workshop, and supply fresh challenges for experienced learners.
These are possibilities to investigate—not outcomes we can claim in advance.
The practical ambition is modest but meaningful: let each round of application contribute something useful to the next.
The workshop ends on the calendar. Its participants’ experience could give the learning a new beginning.
“If this possibility works, what does it make possible next?”
