Gains:
- Ability to attribute the future to several scenarios (optimistic, basic, pessimistic) rather than a single prediction and prepare an answer to each of them in advance
- Ability to reveal blind spots of the plan through pre-mortem and hidden assumption hunting and attach an early warning signal to each scenario
- Ability to strategically decide how much resources to allocate to which future, knowing that artificial intelligence scenarios are not prophecies but thinking tools
The future is uncertain; But uncertainty is no excuse for lack of planning. Good managers don't try to predict the future — because that's impossible — instead they prepare for different futures. Scenario planning — a method of identifying several different possible future situations and preparing a response to each one — is exactly that: “what do we do if this happens?” Asking the question beforehand, not when it comes to work. AI is a very powerful thinking partner in generating scenarios, testing assumptions, and drafting responses for each scenario. Frontier: AI makes scenarios and possibilities visible; How much to prepare for which scenario and how to connect the resource is the manager's strategic decision.
Why multiple scenarios and not just one prediction?
The classic mistake is to pin the future on a single number: "We will grow 15 percent next year." If this prediction does not come true one day, the plan will collapse. However, reality is an interval. Scenario planning therefore establishes at least three worlds: optimistic (if things go well), baseline/expected (most likely course), and pessimistic (if things go wrong). In some cases, a fourth “breakthrough” scenario is also added (major unexpected event: crisis, regulatory change, new competitor). If you prepare a response for each scenario in advance, plan instead of panic will come into play, no matter which way reality turns.
AI is great at generating and enhancing these scenarios: you give an assumption (“if our main supplier increases the price by 30 percent”), it opens up knock-on effects and possible responses. But remember: the scenarios the AI produces are based on the assumptions and historical patterns you provide; It cannot fully predict truly new and unique ruptures (events that have never occurred before). Scenarios are a thinking tool, not a prophecy.
Tip: The real value in scenario planning is not "knowing the right scenario" but "not being caught unprepared." Identify a simple “early warning sign” (if this happens, we are entering this scenario) for each scenario. Set this up by telling the AI to “suggest 2 early signals that will indicate each scenario is occurring.”
Testing assumptions and finding blind spots
Every strategic plan is based on a set of assumptions: "demand will grow", "costs will remain constant", "the competitor will not react". If these assumptions prove quietly wrong, the plan collapses. There is great value in using AI as a “devil's advocate”: handing over your plan and asking “what are the hidden assumptions on which this plan is based, which one turns out to be wrong and the plan will collapse?” ask. AI reveals blind spots and overly optimistic assumptions that you can't see.
This "pre-mortem" approach prevents many disasters. "One year later this plan failed; what were the 5 most likely reasons?" question quickly produces a powerful list of risks with AI. But the risks found are a warning; You decide, with your industry knowledge, which ones are real threats and which ones to allocate resources to.
three mini cases
Case 1 — Three-scenario budget. A medium-sized manufacturer was tying its next year's budget to a single growth figure. This time, the manager set up three scenarios with AI: demand +15% (optimistic), +5% (baseline), -10% (pessimistic). He planned in advance which costs could be cut and which investment would be postponed in each scenario. When the year shifted to the pessimistic side, cost adjustment was made within two weeks, thanks to the ready-made plan; His opponents hesitated for months.
Case 2 — Launch recovered by pre-mortem. A service company was going to launch a new product. The team was excited. Before the launch, the manager told the AI "this launch failed, write down the 5 most likely reasons." In YZ's list was the item "customer support team does not know the new product"; no one had thought of this. Training was provided to the support team before launch. Post-launch support requests were met without any problems.
Case 3 — Supply disruption scenario. One business was dependent on a single imported component. The manager asks "what do we do if this component does not arrive for 3 months?" He worked on his script with AI. AI listed responses such as alternative supplier, stock increase and product redesign by cost/time. The manager pre-agreed with a low-cost safety stock and a backup supplier. When a supply crisis actually occurred six months later, the business continued production.
Four copyable templates
1) Production of three scenarios:
Your role: strategy advisor.Situation: [decision/plan and key assumptions].Generate me 3 scenarios: optimistic, fundamental, pessimistic. For each scenario: triggering condition, key business impacts, 2 recommended preparation steps. Note that these are predictions and not definitive prophecies.
2) Pre-mortem (failure analysis):
I will implement this plan: [plan].Assume this plan FAILS after one year. Write down the 5 most likely reasons for failure, starting with the most critical. For each reason, suggest a precaution that can be taken now. Don't be optimistic; be honest and critical.
3) Hidden hypothesis hunt:
Examine the following strategic plan: [plan].List the HIDDEN assumptions on which this plan is based that are not explicitly written down. For each assumption: the impact on the plan if it turns out to be wrong and a way to test it early. Help me find my blind spots, don't confirm the plan.
4) Early warning signals:
For the following scenarios [scenario list], suggest 2 early warning signals (trackable, tangible indicators) that will indicate that each is starting to happen. Match the table with the table which preparation plan will be activated if any signal is seen.
Weak prompt / Strong prompt
Weak prompt:
What will happen next year?
This is a prophetic request; AI cannot predict the future. Makes general, vague, useless guesses.
Powerful prompt:
Your role: strategy consultant. My business: a regional food distribution company. I'm setting up a budget for next year. I am uncertain about the increase in demand. Generate me 3 scenarios (demand +12%, +3%,-8%). In each scenario, which cost item should I cut and which investment should I postpone? give 2 preparation steps and 1 early warning signal for each. Emphasize that these are guesses.
Size
poor approach
Strong approach
Request type
prophecy
Scenario
Context
None
Net (sector, decision)
output
One vague guess
3 scenarios + preparation
early warning
None
Yes
availability
low
high
Common mistakes
- Connecting the future to a single number. Set up a range and a few scenarios; Don't rely on just one guess.
- Asking for prophecy from AI. AI does not know the future; It produces scenarios and possibilities, not certainty.
- Not questioning hidden assumptions. Uncover blind spots with pre-mortem and hypothesis hunting.
- Writing the script and shelving it. Connect an early warning signal and a ready response to each scenario.
- Allocating equal resources to all scenarios. Prioritize by likelihood and impact; This decision is yours.
Caution: Scenario planning should not create the illusion of control over the future. The scenarios the AI produces are based on past patterns and your assumptions; It cannot truly predict new ruptures (crises that have never been seen before). The aim is not to know the future, but to not be caught unprepared.
In summary
Scenario planning is the way to put a plan in place rather than panic in an uncertain future. Tie the future not to a single number, but to at least three scenarios (optimistic, fundamental, pessimistic) and prepare answers to each of them in advance. Uncover your plan's blind spots with a pre-mortem and hidden assumption hunt. Connect an early warning signal to each scenario. AI is a powerful partner in scenario generation, hypothesis testing, and response drafting; But scenarios are not prophecies, but thinking tools, and it is the manager's strategic decision for which future to prepare and how much.
Application task
Choose a real resolution or plan for the upcoming period. Ask the AI for optimistic/basic/pessimistic scenarios with the “Three scenario generation” template above. Then, with the “Pre-mortem” template, extract the 5 most likely reasons for failure of the plan and write a countermeasure for each. Finally, connect an early warning signal to each scenario and decide how much resources to allocate to which scenario.
checklist
- [ ] Did I attribute the future to several scenarios rather than a single prediction?
- [ ] Have I added a preparation step to each scenario?
- [ ] Have I extracted the reasons for failure with pre-mortem?
- [ ] Have I questioned hidden assumptions and determined a test path?
- [ ] Did I connect an early warning signal to each scenario and prioritize it myself?