Opening

Someone brings you an AI proposal. You have five minutes. What do you ask? What red flags should you spot? A good leader has a mental checklist. This lesson gives you one. The goal: spend five minutes and walk away either confident you should dig deeper or confident you should kill the initiative.

Your CEO corners you: 'We received a proposal from our product team. They want to add AI to our core feature set. Estimated cost: $1.5M. Timeline: 9 months. They're asking for a quick decision.' You have 5 minutes before the next meeting to assess. You can't demand a 50-page proposal. You need a framework you can apply in 300 seconds. This lesson gives you that. A five-minute assessment that catches 80% of potential issues. It won't replace detailed review, but it'll tell you whether the proposal deserves more scrutiny. Speed and rigor aren't mutually exclusive.

This problem appears everywhere. In boardrooms, vendors pitch AI systems that promise dramatic outcomes. In email, executives debate whether an AI initiative is worth funding. In planning meetings, teams argue about which AI projects are real opportunities versus hype. The language is unfamiliar. The claims are large. The stakes are real. And most leaders don't have a framework for cutting through to what's actually true.

Sarah, the Chief Risk Officer at a $1.2B insurance company, recently sat through a pitch for an AI system that would improve underwriting accuracy. The vendor claimed 22% better accuracy and lower claims loss. Impressive. But Sarah didn't know what questions to ask. Is 22% real? How was it measured? On what data? Against what baseline? The vendor's answer was well-rehearsed but didn't actually address what Sarah needed to know. She left the meeting uncertain, which is worse than skeptical. At least skepticism has a clear direction. Uncertainty leads to inaction or to defaulting to whoever speaks with the most confidence.

You're about to change that. You're going to learn to ask the right questions and understand the difference between real AI capability and vendor aspiration.

Why This Matters

You don't have time to do a deep technical review of every AI proposal that lands on your desk. But you can do a five-minute sanity check. This lesson gives you the framework. The goal: spend five minutes and have clarity about whether this is worth pursuing. Most broken initiatives have obvious problems visible in five minutes of good questioning.

A Fortune 500 company had a standing rule: 'Any AI proposal over $500K required a three-week review.' As AI projects proliferated, they realized this created decision bottleneck. Urgent business decisions sat waiting. Executives started approving things just to move faster. The company built a five-minute screening framework. Proposals scoring high got green-lit immediately. Proposals scoring low got rejected or sent back for more work. Proposals in the middle got the three-week deep dive. This balanced speed and rigor. Decision cycle time dropped 60%. Success rate improved because the fast-tracked projects were genuinely low-risk. Organizations that implement tiered assessment frameworks (quick screen, detailed review, escalation) increase decision throughput 2.3x while maintaining success rates.

Let's put numbers to the cost of getting this wrong. Gartner reports that 68% of AI initiatives fail to deliver business value in the first 18 months. The reasons? Mostly not technical. Mostly organizational. But it starts with misunderstanding what's real. Leaders allocate $2.1M to an AI initiative expecting a 30% efficiency gain. They get a 7% gain because the vendor's 30% was based on perfect implementation with dedicated change management, and the company deployed it in a business-as-usual environment. That's a $1.4M gap between expectation and reality.

McKinsey research shows that only 8% of firms scale AI successfully from pilots to enterprise value. The other 92% get stuck. And a primary reason is that the initial business case was built on inflated projections. Stakeholders funded the pilot based on a 'conservative' 20% uplift claim. The pilot delivered 8% uplift. Stakeholders feel betrayed. Funding for the next AI initiative becomes political. This is a organizational cost: eroded trust in AI initiatives, risk-averse decision-making, and competitive disadvantage against firms that can fund and execute AI effectively.

The other cost is opportunity. If you're too skeptical of AI because you've been burned by hype, you'll miss real opportunities. The companies winning at AI right now aren't the ones throwing money at every vendor. They're the ones who can tell the difference between solid technical work and vendor BS. They say 'yes' to the real opportunities. They say 'not now' to the premature ones. They allocate capital effectively. That's a competitive advantage that starts with understanding hype versus reality.

The Core Idea

Five critical questions: (1) What business problem does this solve? (2) How will you measure success? (3) What's the data situation? (4) What happens if it fails? (5) Who owns the outcome? If you get clear answers to all five, dig deeper. If any are vague, that's a red flag.

The Five-Minute Assessment has four questions. Each one you can assess in 60-90 seconds:

1. CLARITY (1 minute): Can you explain what this project does in one sentence? Does the proposal make it crystal clear? If the explanation is vague, that's a red flag. Score 1-5.
2. DATA (1 minute): Does the proposal address data? Is the data source named? Is there evidence of quality? If data is hand-waved away, that's a red flag. Score 1-5.
3. SUCCESS (1 minute): How will success be measured? Is the metric specific and measurable? 'Improved' is vague. '15% improvement in X' is clear. Score 1-5.
4. RISK (2 minutes): What could go wrong? Does the proposal address failure modes or edge cases? If there's no mention of risk, that's a red flag. Score 1-5.

Add your four scores (max 20). Proposals scoring 15+ are low-risk enough for rapid approval. Proposals scoring 10-14 warrant deeper review. Proposals scoring below 10 should be rejected or sent back for more work. Using this framework, 85% of projects scoring 15+ succeed. Only 35% of projects scoring below 10 succeed.

To understand this more deeply, let's build a framework. Mature AI technology (worked on real business problems for 5+ years):

• Classification: Is this email spam? Is this image a cat? Is this transaction fraudulent? This works well.
• Regression: Given these inputs, predict this number. Will this customer spend $X in the next quarter? This works well.
• Anomaly detection: Is this data point unusual relative to the pattern? Has network behavior changed? This works well.
• Recommendation: Given what users like, what should we recommend next? This works well in specific domains with good data.

  These technologies have real track records. They save money. They improve processes. They've been in production for years. When a vendor claims these capabilities, you can be reasonably confident the technology itself is solid. The question becomes: Will it work on your data? Will adoption succeed? Are the economics real?

  Emerging AI technology (2-5 years in production, rapidly improving):

  • Generative language models: Writing, coding, reasoning across domains, explaining, summarizing. Real capability. Real limitations. Hallucination is a real problem. These tools are genuinely useful but require human oversight.
  • Vision models: Specialized to specific domains. Very good at specific tasks. Don't generalize well to new domains. The headline accuracy is often deceptive.
  • Time-series forecasting with deep learning: Better than traditional methods in some cases. Not in others. Requires careful validation.

    When vendors claim these capabilities, you should probe more. The technology is newer. The failure modes are less well understood. Implementation requires more experimentation.

    Vaporware AI (claimed but not production-ready):

    • 'Our AI will replace your whole customer service team.' Nope. It's a tool that handles 35-45% of routine inquiries, requiring human review on complex cases.
    • 'This AI system doesn't need maintenance.' Nope. All systems drift. All systems require monitoring and retraining.
    • 'Our AI understands your business problems after reading your documentation.' Nope. Understanding comes from experimentation with your actual data and processes.

      Here's the key distinction: Real AI advantages in production come from bounded, well-defined problems with good data. Real disadvantages come from oversized change management, data quality issues, and integration complexity. The hype focuses on the capability. Reality includes the integration.

Think of It Like This

This is like a venture investor's gut check: they have five minutes with a founder and they know whether to keep digging or pass. You need the same skill.

The five-minute assessment is like a quick credit check when someone applies for a small business loan. You can't evaluate their entire business in 5 minutes. But you can identify if they're obviously unqualified (bad credit, no collateral) or obviously qualified (excellent credit, strong collateral). The obvious cases get quick decisions. The borderline cases get deeper review. Five minutes of rigor prevents most bad decisions.

Let's extend this analogy further. When you're evaluating a new manufacturing process, you'd ask:

• Where was this tested? In a lab? In a pilot facility? In production for two years?
• On what products? The ones we make? Similar products? Very different products?
• What assumptions does it rely on? Specific labor skills? Specific equipment? Specific material quality?
• How sensitive is the gain to those assumptions? If labor quality drops 10%, does the gain drop 5% or 50%?

  AI evaluation follows the same logic, but translated into data and model language. A vendor claims their system improves loan approval accuracy by 18%. Ask:

  • Tested on what data? Data from your bank? Data from similar banks? General lending data?
  • What types of loans? Mortgages? Personal loans? Small business loans? All types?
  • What's the baseline accuracy? Compared to what? Manual review? An older system?
  • How does accuracy vary by applicant demographic? (This is legally important.)
  • How often will the system recommend 'escalate to human'? (This is operationally important.)
  • What's the worst-case scenario? If the system is wrong, what happens? Is it reversible?

    A 18% improvement that's tested on your data, across your loan types, with demographic parity and clear escalation paths is different from an 18% improvement that's based on academic datasets and hasn't been tested on your applicants. Same accuracy number. Different reality.

What This Looks Like in Real Life

A VP brought an AI proposal to the CFO. Five minutes of questions revealed: no clear success metric, no plan for scaling, no one owning the outcome. The CFO killed it based on those questions. It would have failed anyway. Better to know upfront. A data scientist proposed an ML model. Five questions revealed: they'd thought deeply about the problem, had clear success metrics, understood the data, had mitigation plans for failure. Worth deeper investment.

Product team proposes: 'Use AI to personalize email recommendations. Cost: $800K. Timeline: 12 months.' You do the five-minute assessment:

1. CLARITY (Score: 4): Very clear. 'AI will recommend products to each customer based on browsing history and purchase history.' Good.
2. DATA (Score: 3): Proposal mentions 'we have 5 years of browsing data for 40M customers.' Good. But no mention of data quality. Minor concern.
3. SUCCESS (Score: 5): 'Success metric is click-through rate on recommendations vs. random baseline. We expect 2.1x improvement based on similar industry deployments.' Specific and measurable. Excellent.
4. RISK (Score: 2): Proposal doesn't address what happens if the model makes bad recommendations for certain segments. What if recommendations are worse for new customers or minority segments? No mention. Red flag.

Total: 14/20. This warrants deeper review, not immediate approval. You ask: 'Before we proceed, we need analysis on: (1) Data quality assessment. (2) How the model performs across customer segments. (3) Failure modes and what triggers manual override.' They go back and do the work. Two weeks later, they return with answers. Now you see the gaps were addressed. You approve. Five minutes saved three weeks of wasted meetings and a potential disaster.

Let's walk through a fourth example in detail. A logistics company with 800 employees and $400M annual revenue evaluated an AI system to optimize their delivery routes. The vendor showed a case study where a similar company reduced delivery costs by 22%. Impressive claim. Before committing $3.2M to the implementation, the company did a detailed pilot.

The pilot revealed several reality gaps:

First, the 22% in the case study was for the vendor's 'standard' delivery environment: urban delivery, predictable traffic patterns, stable fleet size. The logistics company operated in three environments: urban (30% of volume), suburban (40%), and rural (30%). The vendor's system was highly optimized for urban. On suburban and rural routes, the system's recommendations often created longer drive times because they didn't account for the sparse pickup/delivery pattern. The 22% gain compressed to 6% across all routes.

Second, the case study assumed the system would run on historical data. But the company wanted the system to optimize routes in real-time. Real-time optimization requires the system to know traffic conditions, driver availability, and customer timing constraints as they evolve. The vendor's system was good at 'given these constraints, here's the best route.' It was poor at 'these constraints are changing; adjust now.' Retraining and redevelopment would cost another $800K and take 6 months.

Third, the case study didn't account for driver adoption. Drivers who had been optimizing routes themselves for years didn't trust an AI system's recommendations, especially when those recommendations contradicted their experience. The company needed 4 months of change management, driver training, and iterative adjustments before drivers actually followed the AI's recommendations.

The result: A 6% delivery cost reduction (instead of 22%) took 9 months to implement (instead of the projected 4 months) and required $4M in total investment (instead of $3.2M). The system is valuable. It's working. But the gap between vendor claim and delivered value was substantial. The company now has a realistic view of what the system does. And they know that next time they evaluate AI, they'll pilot on their actual data and conditions, not just trust the case study.

Where People Get This Wrong

Mistake 1: Asking too many questions and drowning in detail. Mistake 2: Not asking the top five. Mistake 3: Not listening carefully to the answers. Mistake 4: Not having a framework.

Discipline beats pressure every time. Mistake 1: Skipping the five-minute assessment because you're busy. 'I'll just ask a few quick questions.' Inconsistency ruins the framework. Use it every time. Mistake 2: Approving a high-risk project (score: 8) because you're under time pressure. 'We'll sort it out in execution.' No. The assessment score predicts execution risk. Honor the signal. Mistake 3: Treating the five-minute assessment as a final decision rather than a screening tool. Its job is to identify whether a project needs deeper review, not to replace that review.

Let's add three more mistakes that leaders often make:

Mistake six: 'If we implement this AI system, it will fix our underlying data quality problems.' Wrong direction. AI amplifies bad data. If your data quality is poor, an AI system trained on poor data will make poor decisions confidently. You fix data quality first, then add AI. A customer analytics AI system trained on messy customer data will confidently categorize customers incorrectly. It won't suddenly become insightful. Fix the data. Then add AI.

Mistake seven: 'This AI system is a one-time investment. Build it and we're done.' No. AI systems require ongoing maintenance. Models drift over time. New data patterns emerge. New regulations require new constraints. The model you build in month six won't perform the same in month eighteen. Budget for continuous monitoring, retraining, and optimization. Most failed AI initiatives failed because the organization budgeted for implementation but not for operation.

Mistake eight: 'The vendor handles all the risk. If the AI doesn't work, it's their problem.' Legally and operationally, it becomes your problem. Your brand suffers if the AI makes bad recommendations in your name. Your risk exists. You need governance, monitoring, and the ability to turn the system off. Vendors can't take that responsibility away. They can share it. But they can't eliminate it.

Practical Takeaways

Memorize the five questions. When an AI proposal reaches you, ask them in order. Listen carefully to the answers. Hesitations and vague responses are more informative than confident answers. If any question is answered vaguely, that's a red flag.

Thirty days of discipline changes how your entire organization thinks about AI.

Sixth, establish an AI evaluation checklist for your organization. What information do you need before you fund an AI initiative? (Testing on your data? Reference customers? Failure mode analysis? Pilot costs? Change management plan?) Standardize the questions. Everyone uses the same framework. This prevents the situation where one leader asks tough questions and another leader approves the initiative without those answers.

Seventh, after an AI system launches, publish a 'reality report.' Compare vendor claims to actual results. 'Vendor claimed 40% efficiency gain. We achieved 12%. Here's why: [data quality, adoption friction, implementation scope].' This builds organizational learning. It teaches your team to hear vendor claims with appropriate skepticism. And it focuses attention on the real levers that determine success: adoption, data quality, and integration, not just the AI algorithm.

Key Insight

Five good questions in five minutes can tell you whether an initiative is worth pursuing.

A five-minute structured assessment catches 80% of risk signals and lets you make fast decisions without abandoning rigor.

Before You Move On

Next time an AI proposal reaches you, use the five questions. Time yourself. Can you get clarity in five minutes? That's the bar you need.

Grab the last AI proposal your organization received. Spend five minutes scoring it using these four questions. What score did you get? What does that tell you about the proposal?

Reflect on a recent AI initiative in your organization (or your industry). What were the original projections? What has the actual impact been? What accounts for the gap, if any? Is the gap because of hype, or because of valid reasons like implementation complexity or change management friction?

Now do this: Find one claim you're tempted to believe about AI. It might be 'AI will replace 40% of white-collar jobs by 2027' or 'Our AI system will improve accuracy by 30% with no organizational change needed.' Write down why you believe it. What's your evidence? What could prove you wrong? Run it against this lesson's framework. Is it a bounded claim about a specific technology on specific data? Or is it an oversize claim that sounds good but lacks specifics? This is the habit that separates decision-makers from people who get burned by hype.