This section defines the core collaboration pattern that governs how human professionals and AI Knowledge Workers work together inside Cyrenza. In a hybrid workforce, performance gains depend on more than task delegation. They depend on a repeatable interaction structure that preserves human control, strengthens quality, and prevents unreviewed outputs from becoming operational decisions.
Cyrenza operationalises collaboration through a disciplined loop that standardises how work is produced and approved. The sequence is designed to ensure that AI Knowledge Workers contribute execution capacity while the accountable professional retains oversight and decision authority. Learners learn the Propose, Review, Refine, Decide loop as the default pattern for professional workflows, particularly in environments where accuracy, governance, and defensibility are required.
The loop begins with AI-driven execution that produces a structured draft, plan, or option set aligned to the task objective and constraints. The human professional then reviews the output for factual accuracy, reasoning quality, tone, and alignment with organisational standards. Feedback is treated as operational instruction, enabling the AI Knowledge Worker to refine the work product with precision. The cycle concludes with a human decision step, where the professional approves the deliverable and authorises its use.
3.1 The Propose, Review, Refine, Decide Loop
3.1.1 Purpose of the Loop
Cyrenza applies a deliberate interaction pattern to ensure that collaboration with AI Knowledge Workers remains governed by human control. In professional environments, the main risk is not that an output is imperfect. The main risk is that an imperfect output is treated as final work without adequate review. A repeatable collaboration structure prevents this risk by formalising how work moves from first draft to approved deliverable.
The Propose, Review, Refine, Decide loop is designed to achieve three outcomes:
-
Increase execution speed without weakening professional responsibility
-
Improve quality through structured human feedback and iteration
-
Preserve decision authority and accountability with the human professional
This loop functions as a workflow discipline. It is the default pattern for producing outputs that are suitable for organisational use, governance review, and stakeholder decision-making.
Hybrid work becomes unreliable when collaboration is informal. Informal collaboration produces inconsistent outputs because objectives, constraints, and review standards drift from one task to the next. Cyrenza standardises collaboration to make output quality repeatable and review steps predictable.
A standard pattern also reduces operational friction. Teams know what to expect at each stage. Outputs become easier to review because structure remains consistent. Responsibility remains clear because the approval step is explicit.
3.2 Stage One: AI Proposes
3.2.1 What “Propose” Means
In the propose stage, the AI Knowledge Worker produces an initial work product aligned to the objective and constraints provided by the human professional. The output may take the form of:
-
A structured draft such as a memo, brief, report, or email
-
A plan such as a workflow, project outline, process map, or campaign structure
-
An option set such as scenarios, alternatives, negotiation positions, or decision pathways
-
An analytical output such as variance drivers, risk flags, segmentation analysis, or clause extraction
The propose stage is designed for speed and coverage. It allows the organisation to move quickly from a blank page to a structured foundation.
3.2.2 Output Requirements in the Propose Stage
A proposal is most useful when it is review-ready. Cyrenza proposals should be structured in a way that supports rapid human evaluation. This typically includes:
-
Stated objective and scope of the output
-
Assumptions and constraints used
-
Core findings or recommendations clearly separated from supporting detail
-
Open questions and uncertainties explicitly flagged
-
Suggested next steps or validation requirements
3.3 Stage Two: Human Reviews
3.3.1 The Role of Human Review
Human review is the control mechanism that preserves accountability. The responsible professional assesses whether the output is accurate, appropriate, and aligned to organisational standards. Review converts a draft into a professional work product through evaluation, correction, and prioritisation.
Review also functions as risk management. It identifies where assumptions are weak, where evidence is missing, and where outputs require refinement before use.
3.3.2 Review Dimensions
Learners are trained to review outputs across four dimensions.
Accuracy and Evidence Alignment
-
Are factual claims correct and supported by available evidence
-
Are calculations, comparisons, and extractions accurate
-
Are uncertainties clearly stated rather than implied
Reasoning Quality and Assumption Discipline
-
Are assumptions explicit and realistic
-
Are trade-offs presented clearly
-
Are conclusions logically connected to the inputs
Tone, Professional Standards, and Stakeholder Fit
-
Is the tone appropriate for the audience
-
Does the output match internal communication norms
-
Does it maintain the required level of professionalism and clarity
Strategic Alignment and Governance
-
Does the output align with organisational objectives and constraints
-
Does it respect permissions, policies, and approval pathways
-
Does it avoid commitments that exceed authority levels
3.3.3 Review Outputs as Operational Instructions
In Cyrenza, feedback is treated as instruction, not commentary. Effective review feedback includes:
-
What is correct and should remain unchanged
-
What must be corrected and why
-
What must be added or removed
-
What standard or constraint must be applied
-
What format or structure is required for the next iteration
This approach makes refinement faster and reduces repeated cycles.
3.4 Stage Three: AI Refines
3.4.1 What “Refine” Means
Refinement is the structured improvement of the initial work product based on human feedback. The AI Knowledge Worker adjusts content, structure, tone, assumptions, and formatting until the output meets the defined standard.
Refinement is not limited to rewriting. It includes correction, re-structuring, re-calculation, re-synthesis, and re-framing within the boundaries of the task.
3.4.2 Types of Refinement Common in Professional Work
Learners commonly request refinement in the following forms:
-
Correction of factual errors and unsupported claims
-
Strengthening of reasoning, including clearer trade-offs and assumptions
-
Improved structure, including executive summaries and decision-ready framing
-
Alignment with templates and internal standards
-
Tailoring to stakeholder needs, including tone and emphasis adjustments
-
Addition of risk notes, compliance flags, or escalation triggers
3.4.3 The Value of Iteration
Iteration improves quality without restarting work. Each cycle preserves continuity, making the work more stable over time. This continuity reduces rework and improves consistency across teams.
3.5 Stage Four: Human Decides
3.5.1 Decision as a Formal Control Step
The decide stage is where the accountable professional approves the final deliverable and authorises its use. This step ensures that responsibility remains human-owned and that the organisation maintains governance discipline.
Decision authority includes determining whether the output is ready to be used in:
-
Executive decision-making
-
Client-facing communication
-
Legal or regulatory processes
-
Operational execution
-
Reporting and documentation
3.5.2 What Human Approval Includes
Approval typically includes:
-
Confirmation that the output meets the required standard
-
Acceptance of key assumptions and identified trade-offs
-
Acknowledgement of residual uncertainty and risk
-
Confirmation of alignment with policy and permissions
-
Authorisation for distribution or action
3.5.3 Accountability Remains Fixed
The decide stage anchors accountability. The final responsibility for consequences remains with the human role empowered to approve and act.
3.6 Applying the Loop Across Cyrenza Workflows
3.6.1 The Loop as a Universal Collaboration Pattern
Learners should treat the loop as the default pattern for knowledge work across industries:
-
Finance: scenario packs, cash flow updates, risk flags
-
Legal: contract deviation reports, clause extraction, governance summaries
-
Insurance: triage briefs, fraud signals, structured case summaries
-
Marketing: funnel diagnostics, test plans, messaging frameworks
-
Consulting: synthesis maps, option sets, executive briefs
The task changes by domain. The collaboration structure stays stable.
3.6.2 Indicators of Effective Use
The loop is working effectively when:
-
Outputs are produced faster while remaining review-ready
-
Review time decreases because structure becomes consistent
-
Iterations are fewer because feedback is precise
-
Decision ownership is explicit and documented
-
Teams build continuity across workstreams instead of restarting
3.6.3 Indicators of Misuse
Teams may be misusing the loop when:
-
Outputs are treated as final without review
-
Feedback is vague, leading to repeated iterations
-
Decision authority is unclear or unassigned
-
Outputs are not structured for review and governance
-
Teams bypass approval pathways due to urgency