AI for Decision Support: Analyze Options, Risks & Make Better Decisions

```

AI for Decision Support

Decision-making is a fundamental activity in education, business, administration, technology, project management, and everyday life. Many decisions involve multiple alternatives, incomplete information, competing priorities, constraints, and potential risks.

AI can support decision-making by helping users structure a problem, identify options, compare alternatives, analyze criteria, explore scenarios, identify risks, challenge assumptions, and organize available information.

Key Principle:

AI should support human decision-making, not replace the person or authority responsible for the decision. The quality of the decision depends on the quality of the information, criteria, assumptions, and human judgment used in the process.

What Is AI Decision Support?

AI decision support is the use of AI to assist a person or team in analyzing a decision without automatically taking responsibility for the final decision.

Problem → Options → Criteria → Analysis → Risks → Decision → Review

AI can help make the decision process more structured and transparent, particularly when there are several alternatives to consider.

Decision Support vs Automated Decision-Making

AI Decision Support	Automated Decision-Making
Provides analysis	System makes or executes a decision
Human reviews the output	Human involvement may be limited
AI suggests options	System may select an outcome
Human retains responsibility	Accountability requires explicit governance
Useful for exploring alternatives	Useful for predefined, rule-based workflows

Important:

For high-impact decisions involving students, employees, admissions, assessment, finance, employment, discipline, or other sensitive matters, AI output should not be treated as the sole basis for the decision.

Why Use AI for Decision Support?

Decision Challenge	AI Assistance
Too many options	Organize and compare alternatives.
Unclear criteria	Help identify relevant decision criteria.
Complex information	Summarize and structure information.
Hidden assumptions	Challenge assumptions and identify dependencies.
Potential risks	Brainstorm possible risks and mitigation approaches.
Uncertain future	Explore possible scenarios.
Conflicting priorities	Compare trade-offs explicitly.
Need for explanation	Structure the reasoning and decision criteria.

1. Define the Decision Problem

Help me clearly define this
```

decision problem:

[DECISION]

Context:
[CONTEXT]

Objective:
[OBJECTIVE]

Constraints:
[CONSTRAINTS]

Stakeholders:
[STAKEHOLDERS]

Identify:

* The actual decision to be
  made
* Important questions
* Relevant criteria
* Constraints
* Assumptions
* Missing information

Do not recommend a solution yet.

```

2. Identify Decision Criteria

Help identify decision criteria
```

for:

[DECISION]

Context:
[CONTEXT]

Generate relevant criteria
such as:

* Cost
* Quality
* Time
* Impact
* Risk
* Feasibility
* User experience
* Scalability
* Sustainability

Explain why each criterion
may matter.

Do not assign scores yet.

```

3. Identify Available Options

We need to make this decision:
```

[DECISION]

Current options:

[OPTIONS]

Generate additional
realistic alternatives.

Consider:

* Existing options
* Low-cost options
* Technology-based options
* Incremental approaches
* Completely different approaches

Do not assume that any option
is automatically superior.

```

4. Compare Multiple Options

Compare these options:
```

[OPTIONS]

Decision criteria:

[CRITERIA]

Create a comparison table
showing:

Option
Strengths
Weaknesses
Cost
Time
Risk
Feasibility
Expected benefit

Identify information that is
missing or uncertain.

```

5. Create a Decision Matrix

Create a weighted decision
```

matrix for:

[DECISION]

Options:
[OPTIONS]

Criteria:
[CRITERIA]

Assign weights only after
explaining why each criterion
matters.

Use a 1–5 scoring scale.

Show:

Criterion
Weight
Option Scores
Weighted Score
Total

Clearly state that scores are
judgment-based unless supported
by objective data.

```

6. Analyze Cost vs Benefit

Analyze these options:
```

[OPTIONS]

Compare:

* Expected benefits
* Direct costs
* Indirect costs
* Implementation effort
* Maintenance effort
* Risks
* Opportunity costs

Create a structured
cost-benefit comparison.

Do not invent financial
figures. Identify data that
must be supplied.

```

7. Analyze Trade-Offs

Analyze the trade-offs between
```

these options:

[OPTIONS]

Criteria:

[CRITERIA]

Identify where improving one
factor may negatively affect
another.

For example:

Cost vs Quality
Speed vs Accuracy
Flexibility vs Standardization
Innovation vs Risk

Explain the major trade-offs.

```

8. Identify Risks

Analyze the risks associated
```

with this decision:

[DECISION]

Options:
[OPTIONS]

Identify:

* Operational risks
* Financial risks
* Technical risks
* Security risks
* Privacy risks
* People risks
* Implementation risks
* Reputation risks

For each risk provide:

Likelihood
Potential impact
Possible mitigation

Clearly distinguish
hypotheses from known facts.

```

9. Challenge Assumptions

Challenge the assumptions
```

behind this decision:

[DECISION]

Current reasoning:
[REASONING]

Identify:

* Explicit assumptions
* Hidden assumptions
* Weak assumptions
* Dependencies
* Information gaps

For each important assumption
suggest how it could be
validated.

```

10. Act as a Devil's Advocate

Act as a devil's advocate.
```

Decision:
[DECISION]

Preferred option:
[OPTION]

Reasoning:
[REASONING]

Challenge the proposal.

Identify:

* Strong objections
* Risks
* Weak reasoning
* Counterexamples
* Unintended consequences
* Alternative interpretations

Then identify which concerns
require further evidence.

```

11. Analyze Different Stakeholder Perspectives

Analyze this decision from the
```

perspective of:

[DECISION]

Stakeholders:

* Students
* Teachers
* Parents
* School leadership
* IT team
* Finance team
* Management

For each stakeholder identify:

* Priorities
* Benefits
* Concerns
* Risks
* Likely objections

Do not assume that all
stakeholders have the same
priorities.

```

12. Analyze a Decision from Multiple Perspectives

Analyze this decision using
```

these perspectives:

[DECISION]

1. Financial
2. Operational
3. Technical
4. User
5. Strategic
6. Ethical
7. Risk
8. Long-term

For each perspective identify
the most important
considerations.

```

13. Explore Best-Case and Worst-Case Scenarios

Explore possible scenarios for:
```

[DECISION]

Analyze:

1. Best-case scenario
2. Expected scenario
3. Worst-case scenario

For each identify:

* Conditions
* Potential outcomes
* Risks
* Early warning indicators
* Possible response

Clearly label scenarios as
hypothetical.

```

14. Scenario Analysis

Perform scenario analysis for:
```

[DECISION]

Current situation:
[CURRENT STATE]

Create three plausible
scenarios:

Scenario A:
[DESCRIPTION]

Scenario B:
[DESCRIPTION]

Scenario C:
[DESCRIPTION]

For each analyze:

Impact
Risks
Opportunities
Required response

```

15. Analyze "Do Nothing" as an Option

Analyze "do nothing" as a
```

decision option.

Problem:
[PROBLEM]

Other options:
[OPTIONS]

Evaluate:

* Benefits of maintaining
  the current situation
* Costs of inaction
* Risks of inaction
* Opportunities lost
* Future consequences

Compare inaction with the
alternative options.

```

16. Identify the Cost of Inaction

Analyze the possible cost of
```

not addressing this issue:

[PROBLEM]

Consider:

* Time
* Money
* Productivity
* Quality
* User experience
* Risk
* Future opportunities

Do not invent numerical
estimates. Identify what data
would be needed.

```

17. Identify Missing Information

Before making this decision:
```

[DECISION]

Identify the most important
missing information.

Rank the information gaps by:

1. Decision impact
2. Urgency
3. Ease of obtaining
4. Risk if unknown

Suggest how each gap could be
resolved.

```

18. Determine What Evidence Is Needed

For this decision:
```

[DECISION]

Identify the evidence needed
to make a reliable decision.

Separate:

* Facts already available
* Facts that need verification
* Assumptions
* Opinions
* Data required
* External evidence required

Do not fabricate evidence.

```

19. Analyze Data for a Decision

Analyze this data to support
```

a decision:

[DATA]

Decision:
[DECISION]

Identify:

* Important patterns
* Trends
* Outliers
* Comparisons
* Possible explanations
* Limitations

Do not infer causation unless
the data supports it.

State what additional analysis
may be required.

```

20. Ask AI to Explain Data Clearly

Explain this data for a
```

decision-maker:

[DATA]

Audience:
[AUDIENCE]

Provide:

* Key findings
* What the data suggests
* What the data does not prove
* Important limitations
* Decision implications

Avoid overstating conclusions.

```

21. Support a School Leadership Decision

Act as a school-management
```

decision-support assistant.

Decision:
[DECISION]

Context:
[CONTEXT]

Options:
[OPTIONS]

Constraints:
[CONSTRAINTS]

Stakeholders:
[STAKEHOLDERS]

Analyze:

* Educational impact
* Operational impact
* Financial considerations
* Technology considerations
* Staff impact
* Student impact
* Parent impact
* Risks
* Implementation requirements

Do not make the final decision.
Provide a structured analysis.

```

22. Support an Academic Decision

Analyze this academic
```

decision:

[DECISION]

Class / Grade:
[CLASS]

Learning Objective:
[OBJECTIVE]

Options:
[OPTIONS]

Compare:

* Learning impact
* Student engagement
* Teacher workload
* Resources
* Accessibility
* Assessment implications
* Implementation effort

Identify the strongest
trade-offs.

```

23. Support an Examination Decision

Provide decision support for:
```

[EXAMINATION DECISION]

Context:
[CONTEXT]

Options:
[OPTIONS]

Analyze:

* Accuracy
* Confidentiality
* Operational feasibility
* Timeline
* Resource requirements
* Risk
* Verification requirements

Do not invent examination
rules or official requirements.
Identify what must be verified
against authoritative sources.

```

24. Support an IT Decision

Provide decision support for
```

this IT decision:

[DECISION]

Current Environment:
[ENVIRONMENT]

Options:
[OPTIONS]

Compare:

* Functionality
* Performance
* Security
* Privacy
* Compatibility
* Scalability
* Cost
* Maintenance
* Vendor dependency
* Implementation effort

Identify technical information
that still needs verification.

```

25. Support a Software Selection Decision

Help compare these software
```

options:

[SOFTWARE OPTIONS]

Requirements:
[REQUIREMENTS]

Evaluate:

* Required features
* Ease of use
* Integration
* Security
* Privacy
* Scalability
* Support
* Cost
* Implementation
* Data portability

Create a weighted decision
matrix.

Do not assume product features
without verified information.

```

26. Support a Vendor Selection Decision

Compare these vendors:
```

[VENDORS]

Requirements:
[REQUIREMENTS]

Criteria:

* Capability
* Reliability
* Support
* Cost
* Implementation
* Security
* Scalability
* References
* Contract considerations

Identify what should be
verified before selection.

Do not fabricate vendor
information.

```

27. Support a Project Decision

Analyze this project decision:
```

[DECISION]

Project:
[PROJECT]

Current Status:
[STATUS]

Options:
[OPTIONS]

Evaluate:

* Scope
* Timeline
* Resources
* Dependencies
* Risks
* Cost
* Quality
* Stakeholder impact

Provide a decision-support
brief rather than a final
decision.

```

28. Compare Build vs Buy

Analyze a build-vs-buy
```

decision.

Requirement:
[REQUIREMENT]

Build option:
[BUILD]

Buy option:
[BUY]

Compare:

* Initial cost
* Ongoing cost
* Time
* Customization
* Maintenance
* Security
* Scalability
* Vendor dependency
* Technical capability
* Long-term flexibility

Identify which assumptions
require evidence.

```

29. Compare Automation vs Manual Process

Compare these approaches:
```

Manual Process:
[PROCESS]

Automated Process:
[AUTOMATION]

Evaluate:

* Time
* Cost
* Accuracy
* Scalability
* User experience
* Maintenance
* Failure modes
* Security
* Implementation effort

Identify where automation
may not be appropriate.

```

30. Support Resource Allocation

Help analyze resource
```

allocation.

Available Resources:
[RESOURCES]

Projects / Priorities:
[PROJECTS]

Constraints:
[CONSTRAINTS]

Compare allocation options
based on:

* Impact
* Urgency
* Strategic value
* Risk
* Effort
* Dependencies

Identify trade-offs and
unfunded priorities.

```

31. Prioritize Tasks

Prioritize these tasks:
```

[TASKS]

Criteria:

* Urgency
* Impact
* Dependency
* Effort
* Risk

Create:

Priority
Task
Reason
Dependency
Recommended sequence

Do not assume that urgency
always means importance.

```

32. Support Strategic Decisions

Provide strategic decision
```

support for:

[STRATEGIC QUESTION]

Current Situation:
[CURRENT STATE]

Goals:
[GOALS]

Constraints:
[CONSTRAINTS]

Options:
[OPTIONS]

Analyze:

* Strategic alignment
* Benefits
* Risks
* Opportunity cost
* Resources
* Long-term consequences
* Reversibility

Present the analysis for
senior leadership.

```

33. Analyze Reversible vs Irreversible Decisions

Classify this decision:
```

[DECISION]

Determine whether it is:

* Easily reversible
* Partially reversible
* Difficult to reverse
* Essentially irreversible

Explain why.

Suggest an appropriate level
of analysis and approval for
the decision.

```

34. Analyze Decision Reversibility

Analyze the reversibility of
```

these options:

[OPTIONS]

For each identify:

* Reversal cost
* Time required to reverse
* Data or dependency issues
* Stakeholder impact
* Potential lock-in

Explain how reversibility
should influence the decision.

```

35. Identify Second-Order Effects

Analyze this decision:
```

[DECISION]

Identify:

1. Direct effects
2. Indirect effects
3. Second-order effects
4. Possible unintended
   consequences

Consider effects on:

* People
* Process
* Technology
* Cost
* Quality
* Future decisions

36. Conduct a Pre-Mortem

Conduct a pre-mortem for this

proposed decision:

[DECISION]

Assume the decision failed
after implementation.

Brainstorm the most plausible
reasons for failure.

Group them into:

* People
* Process
* Technology
* Resources
* Leadership
* External factors

Then suggest preventive
actions.

```

37. Analyze Decision Biases

Review this decision process:
```

[DECISION]
[REASONING]

Identify possible cognitive
biases such as:

* Confirmation bias
* Anchoring
* Availability bias
* Status quo bias
* Sunk-cost bias
* Overconfidence
* Groupthink

Explain how each could affect
the decision.

Do not claim that a bias is
present without sufficient
evidence.

```

38. Analyze the Status Quo

Analyze the status quo for:
```

[DECISION]

Current State:
[CURRENT STATE]

Identify:

* Benefits of current state
* Problems
* Hidden costs
* Risks
* Dependencies
* Reasons people may resist
  change
* Opportunities created by
  maintaining the status quo

Compare it with proposed
alternatives.

```

39. Create a Decision Brief

Create a one-page decision
```

brief.

Decision:
[DECISION]

Context:
[CONTEXT]

Options:
[OPTIONS]

Criteria:
[CRITERIA]

Include:

* Decision required
* Background
* Options
* Key evidence
* Risks
* Trade-offs
* Recommendation if requested
* Decision owner
* Next steps

Clearly distinguish facts,
assumptions, and judgments.

```

40. Create an Executive Decision Memo

Create an executive decision
```

memo for:

[DECISION]

Audience:
[SENIOR LEADERSHIP]

Include:

1. Decision required
2. Executive summary
3. Background
4. Current situation
5. Options
6. Evaluation criteria
7. Evidence
8. Risks
9. Trade-offs
10. Recommendation
11. Decision required
12. Next steps

Do not present assumptions as
facts.

```

41. Explain a Recommendation

Evaluate this recommendation:
```

[RECOMMENDATION]

Supporting reasoning:
[REASONING]

Explain:

* Why it may be appropriate
* What evidence supports it
* What assumptions it depends on
* What risks remain
* What alternatives exist
* What evidence could change
  the recommendation

42. Ask "What Would Change the Decision?"

Analyze this decision:

[DECISION]

Current preferred option:
[OPTION]

Identify the facts,
conditions, or evidence that
could change the decision.

Group them into:

* Critical
* Important
* Minor

Explain what should be
monitored before finalizing
the decision.

```

43. Create a Decision Tree

Create a decision tree for:
```

[DECISION]

Identify:

* Key questions
* Decision points
* Possible answers
* Resulting options
* Escalation points

Use a clear hierarchical
structure.

Do not assume information
that has not been supplied.

```

44. Identify Escalation Points

For this decision process:
```

[PROCESS]

Identify situations where the
decision should be escalated.

Consider:

* Financial thresholds
* Risk levels
* Security concerns
* Privacy concerns
* Policy issues
* Legal concerns
* Technical complexity
* Stakeholder impact

Mark each as a proposed
governance consideration,
not an official rule.

```

45. Analyze Decision Confidence

Evaluate the confidence level
```

of this decision:

[DECISION]

Available Evidence:
[EVIDENCE]

Identify:

* Strong evidence
* Moderate evidence
* Weak evidence
* Unknowns
* Assumptions

Then classify the overall
decision confidence as:

High
Moderate
Low

Explain the reasoning.

```

46. Create a Decision Review Plan

Create a review plan for this
```

decision:

[DECISION]

Selected Option:
[OPTION]

Success Criteria:
[CRITERIA]

Define:

* What should be monitored
* Metrics
* Review dates
* Warning indicators
* Corrective actions
* Decision reconsideration
  triggers

Make the plan practical and
measurable.

```

47. Conduct a Post-Decision Review

Conduct a post-decision
```

review.

Original Decision:
[DECISION]

Expected Outcome:
[EXPECTED]

Actual Outcome:
[ACTUAL]

Analyze:

* What worked
* What failed
* Unexpected results
* Incorrect assumptions
* Evidence quality
* Decision-process weaknesses
* Lessons learned

Suggest improvements for future
decisions.

```

48. Create a Decision Log

Create a decision log from:
```

[DECISION INFORMATION]

Use these fields:

Decision ID
Date
Decision
Context
Options Considered
Criteria
Evidence
Decision Owner
Decision
Reason
Risks
Review Date
Outcome

Do not invent missing
information.

```

49. Create a Decision-Making Framework

Design a reusable
```

decision-making framework for:

[ORGANIZATION / DEPARTMENT]

Include:

1. Problem definition
2. Information gathering
3. Options generation
4. Criteria definition
5. Risk analysis
6. Stakeholder analysis
7. Option evaluation
8. Decision authority
9. Approval
10. Implementation
11. Monitoring
12. Review

Keep human accountability
explicit at every important
stage.

```

50. Master Decision-Support Prompt

Act as an expert decision
```

support analyst, strategic
advisor, risk analyst, and
critical-thinking partner.

I need support with this
decision:

[DECISION]

Context:
[CONTEXT]

Objective:
[OBJECTIVE]

Options Currently Known:
[OPTIONS]

Stakeholders:
[STAKEHOLDERS]

Constraints:
[CONSTRAINTS]

Available Evidence:
[EVIDENCE]

Current Preference:
[PREFERRED OPTION]

Instructions:

1. Clearly define the actual
   decision that needs to be
   made.

2. Identify missing information.

3. Separate facts, assumptions,
   opinions, hypotheses, and
   unknowns.

4. Identify relevant decision
   criteria.

5. Generate additional
   realistic alternatives.

6. Include the "do nothing"
   option where appropriate.

7. Compare all meaningful
   options.

8. Analyze benefits,
   limitations, costs, effort,
   dependencies, and risks.

9. Identify stakeholder
   perspectives.

10. Analyze short-term and
    long-term consequences.

11. Identify direct, indirect,
    and second-order effects.

12. Explore plausible scenarios.

13. Challenge important
    assumptions.

14. Act as a devil's advocate
    against the preferred
    option.

15. Identify possible cognitive
    biases in the reasoning.

16. Identify evidence that could
    change the decision.

17. If quantitative data is
    available, analyze it
    carefully.

18. Do not invent statistics,
    costs, facts, policies,
    regulations, product
    capabilities, or evidence.

19. Clearly identify
    uncertainty.

20. Use a decision matrix when
    appropriate.

21. Explain trade-offs rather
    than hiding them.

22. Identify quick, reversible
    actions where appropriate.

23. Identify decisions that
    require additional
    authority or expert review.

24. Provide a monitoring and
    post-decision review plan.

25. Keep final decision
    authority with the
    responsible human
    decision-maker.

Return:

A. Decision Statement

B. Context

C. Key Questions

D. Available Evidence

E. Missing Information

F. Assumptions

G. Decision Criteria

H. Options

I. Option Comparison

J. Risk Analysis

K. Stakeholder Analysis

L. Trade-Off Analysis

M. Scenario Analysis

N. Devil's-Advocate Review

O. Potential Biases

P. Decision Matrix

Q. Evidence Needed

R. Recommended Next Steps

S. Monitoring Plan

T. Post-Decision Review Plan

Do not make the final decision
unless I explicitly ask for a
recommendation.

```

Decision-Support Framework

Define → Gather Evidence → Generate Options → Set Criteria → Analyze → Decide → Monitor → Review

A structured decision process reduces the likelihood of making a decision simply because one option appears attractive at first.

AI can contribute at several stages, but the quality of its analysis depends heavily on the quality and completeness of the information provided.

AI's Role in the Decision Process

Stage	Possible AI Role	Human Responsibility
Problem Definition	Clarify and structure the problem.	Confirm the actual problem.
Information Gathering	Organize supplied information and identify gaps.	Verify evidence and sources.
Option Generation	Suggest alternatives.	Determine realistic options.
Criteria	Suggest possible evaluation criteria.	Determine which criteria matter.
Analysis	Compare options and identify trade-offs.	Assess whether analysis is appropriate.
Risk Analysis	Identify possible risks and scenarios.	Assess actual risk.
Recommendation	Provide a structured recommendation when requested.	Accept, reject, modify, or seek further evidence.
Final Decision	Support reasoning.	Make and own the decision.
Review	Analyze outcomes and lessons learned.	Determine corrective action.

Decision Matrix

A decision matrix provides a structured way to compare alternatives against multiple criteria. Criteria can be weighted according to their importance.

Option	Cost	Impact	Risk	Feasibility	Overall Assessment
Option A	Evaluate	Evaluate	Evaluate	Evaluate	Compare using defined criteria.
Option B	Evaluate	Evaluate	Evaluate	Evaluate	Compare using defined criteria.
Option C	Evaluate	Evaluate	Evaluate	Evaluate	Compare using defined criteria.

A numerical score does not automatically make a decision objective. Scores may contain subjective judgments, incomplete information, or assumptions. The criteria and scoring method should therefore be transparent.

Example — School Technology Decision

Situation

A school is considering different technology solutions for a particular operational requirement.

AI can help structure the decision around functionality, integration, security, privacy, cost, support, scalability, implementation effort, and user experience.

Requirement → Options → Criteria → Evidence → Evaluation → Pilot → Review

The final selection should be based on verified product information, institutional requirements, procurement processes, and appropriate human approval.

Example — School Leadership Decision

Situation

School leadership must decide between several approaches to address an operational or academic challenge.

AI can organize the decision into stakeholder impact, educational value, operational feasibility, resource requirements, risks, implementation complexity, and expected outcomes.

Leadership can then use the structured analysis as an input to discussion and decision-making.

Example — Examination Process Decision

Situation

An examination team is evaluating alternative approaches for improving an examination workflow.

AI can help compare alternatives according to confidentiality, accuracy, operational feasibility, timelines, resource requirements, verification, and risk.

Problem → Alternatives → Risk → Verification → Approval → Implementation

Official examination requirements must be checked against applicable authoritative sources rather than inferred by AI.

Example — IT Build vs Buy Decision

Situation

An organization needs a software solution and is considering developing it internally or purchasing an existing solution.

AI can help structure the comparison around customization, development time, total cost of ownership, maintenance, scalability, security, vendor dependency, integration, and technical resources.

Actual costs, capabilities, contracts, security controls, and technical specifications require independent verification.

High-Impact Decisions Require Greater Human Oversight

Decision Area	Recommended AI Role	Human Oversight
Routine planning	Generate and compare options.	Review and select.
Resource allocation	Analyze priorities and trade-offs.	Approve allocation.
Technology selection	Structure comparison.	Verify specifications and approve.
Academic planning	Suggest approaches and analyze options.	Teacher / academic leadership review.
Student-related decisions	Assist with organizing information where appropriate.	Qualified human decision-makers.
Employment decisions	Limited analytical support where appropriate.	Responsible authorized decision-makers.
Financial decisions	Structure scenarios and comparisons.	Verify financial information and approve.
Security decisions	Identify possible risks and alternatives.	Qualified security / technical review.

Common AI Decision-Support Mistakes

Mistake	Better Practice
Asking AI "What should I do?" without context	Define the problem, objectives, options, and constraints.
Accepting AI's recommendation automatically	Evaluate the reasoning and evidence.
Using invented or unverified data	Verify important facts and figures.
Ignoring the status quo	Include inaction as an option where appropriate.
Ignoring trade-offs	Make competing priorities explicit.
Using arbitrary scoring	Define transparent criteria and scoring rules.
Ignoring stakeholder perspectives	Analyze relevant stakeholder interests and concerns.
Confusing correlation with causation	Distinguish observed relationships from proven causes.
Ignoring uncertainty	Clearly identify unknowns and assumptions.
Allowing AI to make high-impact decisions independently	Maintain appropriate human oversight and accountability.
Failing to review the decision later	Define monitoring and post-decision review.

Practical Activity 1 — Compare Three Options

Choose a real but low-risk decision and ask AI to compare three alternatives using clearly defined criteria.

Practical Activity 2 — Build a Decision Matrix

Create a weighted decision matrix and examine how changing the weights affects the ranking of alternatives.

Practical Activity 3 — Devil's-Advocate Review

Give AI your preferred option and ask it to challenge the assumptions, weaknesses, and risks behind that choice.

Practical Activity 4 — Stakeholder Analysis

Analyze a school or workplace decision from the perspective of different stakeholders and identify conflicting priorities.

Practical Activity 5 — Scenario Analysis

Select a decision and ask AI to explore best-case, expected, and worst-case scenarios.

Practical Activity 6 — Build vs Buy

Use AI to structure a hypothetical build-vs-buy technology decision and identify which information must be verified.

Practical Activity 7 — Pre-Mortem

Assume a proposed decision has failed and use AI to identify possible causes of failure and preventive actions.

Practical Activity 8 — Identify Decision Biases

Provide AI with a decision rationale and ask it to identify possible cognitive biases or unsupported assumptions.

Practical Activity 9 — Create a Decision Brief

Convert a complex decision into a one-page brief containing the decision required, options, evidence, risks, trade-offs, and next steps.

Practical Activity 10 — Post-Decision Review

After a decision has been implemented, compare the expected and actual outcomes and use AI to identify lessons for future decisions.

Interview Questions

Q1. What is AI decision support?

AI decision support is the use of AI to help structure, analyze, compare, and evaluate information relevant to a decision while retaining human responsibility for the final decision.

Q2. Can AI make decisions?

AI systems can be designed to automate certain decisions, but AI decision support specifically focuses on assisting human decision processes. The appropriate level of automation depends on the context, risk, governance, and applicable requirements.

Q3. What is a decision matrix?

A decision matrix is a structured comparison method in which alternatives are evaluated against defined criteria, sometimes using weighted scores.

Q4. Why should "do nothing" sometimes be considered?

Maintaining the status quo can itself be a meaningful alternative. Comparing it with other options helps reveal the potential cost or risk of inaction.

Q5. What is a trade-off?

A trade-off occurs when improving one decision factor may require accepting a disadvantage in another factor, such as speed versus quality or cost versus functionality.

Q6. What is a pre-mortem?

A pre-mortem assumes that a proposed decision has failed and asks what could have caused the failure. It is used to identify risks before implementation.

Q7. Why is human oversight important?

AI can produce incorrect, incomplete, biased, or contextually inappropriate analysis. Human oversight provides contextual judgment, accountability, and verification.

Q8. Why are decision criteria important?

Decision criteria make it possible to compare alternatives systematically rather than relying entirely on intuition or the first attractive option.

Q9. Can AI-generated scores make a decision objective?

No. Numerical scores may still depend on subjective judgments, assumptions, and incomplete information. The criteria and scoring methodology must be transparent.

Q10. What is the most important principle of AI decision support?

Use AI to improve the quality and structure of human reasoning, while keeping appropriate human accountability for the final decision.

Examination MCQs

Q1. What is the primary purpose of AI decision support?

Replace every human decision
Support analysis and decision-making
Remove all uncertainty
Guarantee successful outcomes

Answer: B

Q2. What is a decision matrix used for?

Creating images
Comparing alternatives against criteria
Writing code
Sending emails

Answer: B

Q3. Why should assumptions be identified?

To hide uncertainty
To make assumptions appear factual
To determine what needs validation
To eliminate alternatives

Answer: C

Q4. What is a trade-off?

A guaranteed benefit
A conflict between competing factors
A type of AI model
A document format

Answer: B

Q5. What does a pre-mortem examine?

Why a successful project succeeded
How a proposed decision might fail
How to format a report
How to generate images

Answer: B

Q6. Which should be considered when evaluating alternatives?

Only cost
Only speed
Relevant decision criteria and trade-offs
Only AI's preference

Answer: C

Q7. What should happen when AI lacks important information?

AI should invent it
The information gap should be identified
The decision should be automatically rejected
The information should be ignored

Answer: B

Q8. Which is an appropriate use of AI in decision support?

Comparing alternatives against defined criteria
Inventing evidence
Hiding risks
Replacing all decision-makers

Answer: A

Q9. Why should decision outcomes be reviewed?

To identify lessons and improve future decisions
To prove AI was always correct
To remove accountability
To eliminate all future decisions

Answer: A

Q10. Who should normally retain accountability for an important organizational decision?

The AI model
The software vendor
The responsible human decision-maker or organization
The prompt template

Answer: C

Key Terms

Term	Meaning
Decision Support	Use of information, analysis, and tools to assist a human decision-maker.
Decision Matrix	A structured method for comparing alternatives against multiple criteria.
Decision Criteria	Factors used to evaluate and compare possible alternatives.
Trade-Off	A situation where gaining an advantage in one area involves accepting a disadvantage in another.
Scenario Analysis	Exploration of possible future situations and their potential consequences.
Risk Analysis	Identification and evaluation of possible risks associated with a decision.
Pre-Mortem	An analysis that assumes a proposed plan has failed and explores possible causes.
Stakeholder	A person or group affected by, involved in, or interested in a decision.
Assumption	A condition accepted temporarily as true for reasoning or planning.
Evidence	Information or data used to support or evaluate a conclusion or decision.
Decision Bias	A systematic tendency that can influence judgment and decision-making.
Opportunity Cost	The value of an alternative opportunity that is given up when a choice is made.
Reversibility	The extent to which a decision can be undone and its effects reversed.
Status Quo	The existing state or current way of doing something.
Decision Log	A record of decisions, context, alternatives, reasoning, and outcomes.
Decision Confidence	An assessment of how strongly the available evidence supports a decision.
Human Oversight	Human review, supervision, and accountability over AI-supported processes.

Self-Assessment Checklist

☐ Explain AI decision support.
☐ Distinguish decision support from automated decision-making.
☐ Define a decision problem clearly.
☐ Identify decision criteria.
☐ Generate alternative options.
☐ Compare multiple options.
☐ Create a decision matrix.
☐ Analyze costs and benefits.
☐ Identify trade-offs.
☐ Analyze decision risks.
☐ Challenge assumptions.
☐ Use devil's-advocate analysis.
☐ Analyze stakeholder perspectives.
☐ Explore best-case and worst-case scenarios.
☐ Consider the cost of inaction.
☐ Identify missing information.
☐ Determine evidence required for a decision.
☐ Analyze data for decision support.
☐ Support school leadership decisions.
☐ Support academic and examination decisions.
☐ Support IT and technology decisions.
☐ Analyze build-vs-buy decisions.
☐ Analyze automation-vs-manual decisions.
☐ Prioritize resources and tasks.
☐ Analyze reversible and irreversible decisions.
☐ Identify second-order effects.
☐ Conduct a pre-mortem.
☐ Identify possible decision biases.
☐ Create decision briefs and decision logs.
☐ Create decision review plans.
☐ Apply human oversight and accountability.

Key Takeaway

AI can improve decision-making by helping people define problems, generate alternatives, compare options, analyze risks, explore scenarios, challenge assumptions, and structure evidence.

Problem → Evidence → Options → Criteria → Analysis → Human Decision → Review

A strong AI-supported decision process does not simply ask, "What should I choose?" Instead, it asks: "What are my options, what evidence supports them, what are the trade-offs, what could go wrong, and what information is still missing?"

In education, AI can support academic planning, school leadership, examination processes, resource allocation, technology selection, project planning, teacher development, and operational improvement.

In professional environments, AI can support strategic planning, vendor selection, software selection, project decisions, resource allocation, process improvement, risk analysis, and business planning.

However, AI output can be incomplete, incorrect, biased, or based on unsupported assumptions. Important decisions therefore require appropriate evidence, domain expertise, human review, governance, and accountability.

Always remember: AI is a decision-support tool—not the accountable decision-maker.

```