# AI Opportunity Discovery

> AI Opportunity Discovery is a self-paced course by Vin Vashishta, with 14 sections and 70 lessons, that teaches a complete system for finding, qualifying, sizing, and prioritizing the AI opportunities that pay.

URL: https://www.datascience.vin/course-opportunity-discovery.html  
Format: Self-paced  
Price: $295  
Enroll: https://learn.datascience.vin/purchase?product_id=6574180  
Updated: January 2027

- **Format:** On-demand video + exercises
- **Content:** 14 sections · 70 lessons
- **Access:** 1 year, with office hours
- Top-down and bottom-up discovery frameworks
- Feasibility assessments and a reusable estimation framework
- Office hours and email support
- Optional 1:1 sessions with Vin

## Why this course

Most AI initiatives fail before execution starts, in how the opportunity was chosen. As the barriers to building fall, the valuable question moves from how to build to what to build. This course builds the Disruptor's Mindset for discovery, and treats it as a growth function first.

## The shift

| Old pattern | Disruptor's Mindset |
|---|---|
| Collect use cases | Run a pipeline and re-prioritize every quarter |
| Use AI by default | Use the cheapest technology that delivers the outcome, and AI where complexity or uncertainty justify it |
| Answer margin pressure with price increases and layoffs | Answer it with growth |
| Commit first, check feasibility later | Explore the problem, data, and solution spaces before any commitment |
| Give a single-number estimate | Give three bands, with a floor strong enough to carry the initiative |
| React to disruptions | Name the broken assumption and get there first |

**If you're technical: Get into the room where the year's work is chosen.** Learn to qualify, size, and defend opportunities in business terms, so you shape the roadmap instead of inheriting it.

**If you're on the business side: Qualify AI from first principles.** AI manages complexity and reduces uncertainty better than any prior technology. That test needs no technical content, and it tells you when a cheaper tool will do.

## Problems this course solves

### Problems you're facing

- **You're not in the room where it's decided.** Discovery is where the year's work gets chosen. If you aren't in it, you inherit the results, and reopening the decision later costs you credibility.
- **You get handed initiatives you know won't deliver.** Someone else's disconnected KPI lands on your roadmap. When it produces nothing, the technology team absorbs the blame.
- **You're told how to do your job.** “Do this with AI” is an executive specifying your architecture. The course's diagnosis: leadership doesn't yet trust the technical organization to connect technology to value, and that is fixable.
- **You have no track record yet, and everything depends on having one.** Lose credibility before you've delivered and you don't get the budget or latitude afterward. The course sequences the first year around protecting it.
- **You can't quantify the value of your own work.** Without an estimate you're arguing opinion against opinion. The Opportunity Estimation Framework gives you a defensible range.
- **You're treated as C-level in title only.** CEOs say technology leaders “don't focus on value the way the rest of us do.” It's a positioning problem, and better technology won't solve it.
- **You're the only one pushing back.** Challenging a senior stakeholder alone is career-expensive. Surface consequences in a way that recruits allies.
- **Pushing back too hard gets you routed around.** Say no and someone downstream says yes, and you end up on the hook with less control. Accelerate and redirect instead.
- **You over-advocate because you feel you have no control.** The same feeling is driving the executives' behavior toward you. Recognizing it changes the conversation.
- **You're forced to defend a mediocre initiative.** Build what leadership asked for while holding a pivot plan, because they won't blame themselves for the miss.
- **No one says anything in the session.** Silence is close to universal and rarely sabotage. Restate the mission, prompt with their pain points, and keep ideas in your back pocket.
- **You're too good at it and end up owning everything.** Supply every idea and you lose the ownership transfer that makes the process stick.
- **The ideas you get are all digital use cases.** Reframe instead of rejecting: find the complexity or uncertainty inside them.
- **Someone raises a roadblock and the room stops.** Some objections must be honored. The skill is telling those apart from the ones used to block everything.
- **People leave the session feeling stupid.** Then they don't come back. Every session has to be a positive experience.
- **You can't tell hype from a real opportunity.** Four hype-resistant questions filter the demo-driven enthusiasm without dampening it.
- **You can't translate in either direction.** Business context has to reach technical teams, and monetization has to reach the C-suite. Both are your job.
- **You react to disruptions instead of anticipating them.** A learnable behavior: Pragmatic Futurism, and knowing what to listen for when industry leaders speak publicly.
- **You define problems by prescribing solutions.** The most common requirements failure. Problem space definition keeps it buildable without dictating how.
- **You've never estimated something this uncertain.** Ranges feel like hedging until you notice they're how the C-suite already talks to the street.
- **You want to become the person the CEO pulls for growth.** The move from tactical execution lever to strategic partner is a different job, and it's the one this course trains for.

### Business problems

- **Money is spent on the wrong opportunities because no one sized them first.** Without upfront estimation there's no basis for prioritization, so the portfolio drifts toward whatever is loudest or newest.
- **Bolt-on AI consumes the budget bigger opportunities needed.** Replacing a feature with a more expensive AI version, with no re-engineered workflow and no new monetization, is a net negative.
- **Margin pressure answered with price increases, then layoffs.** The reflex response, and the course's central worked counter-example: answer it with growth.
- **“We just need to do something with AI.”** AI as a checkbox produces no return. The Four Rs: reassurance, root cause, redirect, restatement.
- **Use-case thinking instead of pipeline thinking.** With two or three bets in flight, the business can't walk away from a failing one. The opportunity pipeline.
- **Everything is priority one.** Unbounded wish lists that get missed or routed around. The Queen problem and accelerate-and-redirect.
- **Sunk-cost paralysis.** Too many unproven assumptions baked into a product before launch. Assumption budgeting per release.
- **Expensive technology applied where cheap technology would do.** The cheapest-viable-technology rule, with complexity and uncertainty as the test for AI.
- **Technology in search of a problem.** Products shipped because the capability exists. The AI-in-a-coffee-maker case.
- **Failing to see the true size of a paradigm shift.** Treating a fundamental change as incremental. AWS vs. Azure.
- **Incumbents don't innovate until a startup forces them.** A five-decade pattern, now arriving faster. Discovery is the place to break it.
- **Existential exposure to faster new entrants.** Once a competitor is two to five years ahead and accelerating, catching up stops being possible.
- **No mechanism for seeing disruptions before they're obvious.** Pragmatic Futurism and harvesting paradigms from what industry leaders say publicly.
- **Building what everyone else can build.** Without an information or data advantage, competitors replicate it immediately. The four-point progression.
- **Data given away or left unmonetized.** Data-generating processes treated as exhaust. Data as an Asset, with Reddit, Disney, and Lyft.
- **Technology treated as a cost center.** Executives have never been able to ask the technology organization for growth, so they don't.
- **Executives prescribing technical solutions.** Suppresses creativity and inflates cost. The root cause is trust, and framework certainty is the response.
- **Retrenchment after failed pilots.** Real hallucination problems harden into “none of this works,” forfeiting the value that is available.
- **Prioritization by squeaky wheel.** Challenge the premise and ask for proof. Test for prepaid features and demand market research.
- **Prioritization by coolest job title.** Let other organizations challenge it by surfacing displaced revenue.
- **Estimates disconnected from reality.** Commitments made before anyone checked whether the thing can be built. Opportunity Feasibility Assessments.
- **Initiatives blocked mid-flight by data problems.** Access, silos, and quality discovered months in. Data space exploration finds them in week one.
- **Technical teams buried in an unfiltered request queue.** Two upstream gates, problem space and data space, before anything reaches the solution space.
- **Downstream breakage no one anticipated.** Every major strategy change breaks something. Downstream breakage analysis surfaces it up front.
- **Regulatory paralysis.** One legitimate roadblock generalized into a reason to stop everything. Local AI, simpler compliant technology, and narrow targeting.
- **Shipping products customers aren't prepared for.** Value trapped in a product an unready customer base won't adopt.
- **Innovation with no adoption journey.** If you can't define the adoption journey, the opportunity isn't real. The Metaverse vs. GoPro vs. Snap drones.
- **Customers who can't articulate what they need.** No one asked for a Walkman or an iPhone. Sometimes you have to play the visionary.
- **Arriving too late.** The addressable share left is too small to justify the cost. The fourth top-down question.
- **Partnership theater.** Two companies bolting products together on the assumption revenue follows. Adobe and ChatGPT.
- **Competitors attacking from a business model you have no incentive to adopt.** Adversarial opportunity discovery: sometimes the right output is a response plan.
- **Buzzwords hardened into corporate strategy.** KPIs that connect to nothing real, and parts of the business that need that opacity to survive.
- **Bad initiatives that can't be dislodged.** Once committed to a board, reversing reads as incompetence. Discovery is where redirection reads as “we found something better.”
- **The wrong people get laid off.** Talented staff cut because they were pointed at work that created no value. A discovery failure that surfaces as a headcount decision.
- **Frontline teams can't articulate opportunities.** Literacy and translation work so the people closest to the problems can tell you what they need.
- **Executives don't fund the unglamorous substructure.** They never see what it takes to produce an insight. The Anatomy of an Insight workshop.
- **New frameworks rejected as overhead.** Any process that visibly slows the business gets resisted. Every framework here is lightweight by design.

## Course outline

A complete system for finding, qualifying, sizing, and prioritizing opportunities, from what happens before discovery to what to do when it goes wrong. Every exercise uses your own business, rivals, and strategic goals.

1. Introduction (5 lessons)
2. What Happens Before Opportunity Discovery? (6 lessons)
3. The Opportunity Discovery Frameworks (6 lessons)
4. A New Paradigm Of Monetization (5 lessons)
5. A New Mindset & Understanding Of AI Opportunities (3 lessons)
6. Real World Opportunity Discovery Examples (4 lessons)
7. Pragmatic Futurism (6 lessons)
8. Innovation Opportunities (6 lessons)
9. Rethinking The Business: Data As An Asset (6 lessons)
10. Rethinking The Business: Partnership Monetization Opportunities (3 lessons)
11. The Opportunity Estimation Framework (6 lessons)
12. Opportunity Feasibility Assessments (4 lessons)
13. When Opportunity Discovery Goes Wrong & What To Do About It (7 lessons)
14. Finding Opportunity Paradigms (3 lessons)

## Questions

**What is AI Opportunity Discovery?**

AI Opportunity Discovery is a self-paced course by Vin Vashishta, with 14 sections and 70 lessons, that teaches a complete system for finding, qualifying, sizing, and prioritizing the AI opportunities that pay.

**How much does it cost, and what's included?**

$295 for the full course. It includes top-down and bottom-up discovery frameworks, feasibility assessments and a reusable estimation framework, office hours and email support, optional 1:1 sessions with Vin.

**How long does it take?**

It's self-paced, so you set the schedule. There are 14 sections and 70 lessons, about 7.5 hours of video plus exercises and a capstone applied to your own business. Start immediately after you enroll; access runs for a year, with office hours for questions along the way.

**Do I need a technical background?**

No. Prior exposure to the AI Strategy or AI Product Management courses makes the feasibility material familiar, but everything is taught from scratch, with no technical content required.

**Can I expense it?**

Some students use an employer learning budget. Approval depends entirely on your employer's policy.

**Will it teach me the machine learning?**

No. None of the four core courses teach model evaluation, MLOps, training or serving operations, SRE, or security implementation. They teach the other half of the job: deciding what is worth building, proving it will create value, pricing it, and getting an organization to act on it.

**Which course should I take?**

If you have to decide what gets built, start with AI Opportunity Discovery. If you own enterprise strategy and need a mandate, take the Data & AI Strategist Certification. If you have the mandate and need a roadmap that makes money, take AI Product Management. If you own a platform P&L, its pricing, and its governance, take AI & Agentic Platform Monetization. The course matcher scores 40 job titles against all four.

## Instructor

Vin Vashishta is the author of From Data to Profit (Wiley) and has applied every framework in this course with clients including Airbus, Walmart, Siemens, and JPMorgan Chase. More than 9,000 professionals in 47 countries have taken his courses.