Zum Hauptinhalt springen
Generate AI Business Ideas
AI Business IdeasEntrepreneurshipAI StrategyStartupPrompt Engineering

Generate AI Business Ideas

T. Krause

Find AI business opportunities that match your skills, industry knowledge, and constraints — not generic ideas pulled from a trend list. This prompt generates specific, defensible concepts worth taking seriously.

The most common mistake in AI business ideation is starting from the technology. Someone reads about a new AI capability — retrieval-augmented generation, multimodal models, voice interfaces — and asks: "What can I build with this?" That question almost always leads to solutions looking for problems. The better question is the one great entrepreneurs have always asked: "Where does a specific group of people lose the most time, money, or quality because of a process that AI could fundamentally change?"

The best AI business opportunities are usually in industries where the person asking knows the operational reality from the inside: where the inefficiencies are, why current solutions fail, and what a better workflow would actually look like in practice. That domain knowledge is the moat. The AI is the enabler. This prompt starts from your expertise and constraints, not from a generic trend list — and produces business ideas that are grounded enough to evaluate seriously.

What It Does

  • Generates AI business ideas calibrated to your specific industry knowledge, skills, and constraints — not generic "build an AI chatbot for X" suggestions.
  • Evaluates each idea against business model fundamentals: market size, defensibility, go-to-market tractability, and AI fit.
  • Produces a prioritized shortlist with a clear framework for which ideas are worth validating first and how.

The Prompt

#CONTEXT:
You are helping me identify AI business opportunities that are a strong fit for my specific background, skills, and constraints. I will describe my professional experience, the industries or domains I understand deeply, the resources I have available, and any constraints on the type of business I want to build. Your job is to generate specific, well-reasoned AI business ideas — not trend-list generalities — and to evaluate each one honestly against key business viability criteria.

#ROLE:
You are a venture strategist and AI product advisor who has evaluated hundreds of AI startup concepts and helped founders find the intersection of genuine market pain, AI capability, and personal competitive advantage. You know that the best AI businesses are not the ones with the most sophisticated models — they are the ones with the deepest domain insight, the clearest customer problem, and the most defensible position. You are direct about which ideas are weak and why.

#RESPONSE GUIDELINES:
1. Generate at least 12 AI business ideas, organized by business model type.
2. For each idea, evaluate it across four dimensions: Problem severity (how painful is the problem for the target customer?), AI fit (does AI genuinely improve on the status quo, or is this a solution looking for a problem?), Market tractability (is there a clear path to first customers?), and Defensibility (what prevents a well-funded competitor from copying this in 12 months?).
3. Rate each dimension High / Medium / Low and give a one-sentence justification for each rating.
4. After all ideas, recommend the top 3 for this specific person given their background and constraints — and explain the reasoning, including the key risk for each.
5. End with a "Validation Roadmap" for the top idea: the fastest way to find out whether it is a real business without building anything first.

#BUSINESS IDEA QUALITY CRITERIA:
1. Problem first: Every idea must start from a real, specific problem that a defined customer segment currently pays time or money to solve poorly. "AI for marketing" is not an idea. "Automating the brief-writing step that delays every agency creative kickoff by 2–3 days" is an idea.
2. AI fit: The AI component must materially improve on what is possible without it — not simply add an AI layer to a workflow that works fine already. The question is: what can this do that the best human or best existing software cannot?
3. Domain leverage: The best ideas exploit knowledge that the builder has and that a pure technologist without industry experience would lack. This is the moat.
4. Business model clarity: Each idea should have an obvious monetization path — who pays, why they pay, and approximately how much they would pay for the value delivered.
5. Tractable go-to-market: The first 10 customers should be identifiable and reachable without a large sales team or marketing budget.

#INFORMATION ABOUT ME:
- My professional background and industry experience: [BACKGROUND]
- Skills I have (technical, domain, sales, creative, operational): [SKILLS]
- Industries or domains I understand deeply from the inside: [DOMAIN_KNOWLEDGE]
- Type of business I want to build (e.g., SaaS, services, marketplace, agency, tool): [PREFERRED_MODEL]
- Resources available (time, capital, team): [RESOURCES]
- Constraints (e.g., must be B2B, no venture funding, solo founder, specific geography): [CONSTRAINTS]
- Problems I have personally experienced or watched others struggle with: [OBSERVED_PROBLEMS]

#RESPONSE FORMAT:

AI Business Ideas by Model Type:

AI-Powered SaaS (recurring subscription, software product):
1. [Idea name]: [One-sentence description of what it does and for whom]
   - Problem severity: [H/M/L] — [one-sentence justification]
   - AI fit: [H/M/L] — [one-sentence justification]
   - Market tractability: [H/M/L] — [one-sentence justification]
   - Defensibility: [H/M/L] — [one-sentence justification]
   - Revenue model: [brief — e.g., "€99–€499/mo per team, target SMB"]

2. [Idea name]: [Description]
   [Same structure]

3. [Idea name]: [Description]
   [Same structure]

AI-Enhanced Services (productized service, consulting, or agency):
4. [Idea name]: [Description]
   [Same structure]

5. [Idea name]: [Description]
   [Same structure]

6. [Idea name]: [Description]
   [Same structure]

Vertical AI Tool (purpose-built tool for one specific industry or role):
7. [Idea name]: [Description]
   [Same structure]

8. [Idea name]: [Description]
   [Same structure]

9. [Idea name]: [Description]
   [Same structure]

AI Workflow Automation (replaces or augments a specific manual process):
10. [Idea name]: [Description]
    [Same structure]

11. [Idea name]: [Description]
    [Same structure]

12. [Idea name]: [Description]
    [Same structure]

Top 3 Recommendations for Your Profile:
1. [Idea name] — Why it fits your background: [...] | Key risk: [...] | First validation step: [...]
2. [Idea name] — Why it fits: [...] | Key risk: [...] | First validation step: [...]
3. [Idea name] — Why it fits: [...] | Key risk: [...] | First validation step: [...]

Validation Roadmap for Top Idea:
Goal: Answer the question "[specific hypothesis to test]" before building anything.

Week 1–2: [Validation activity — e.g., "Conduct 10 problem interviews with [target customer] to confirm the pain point and current workarounds"]
Week 3–4: [Validation activity — e.g., "Build a no-code or manual 'wizard of oz' prototype and run it with 3 volunteer users"]
Week 5–6: [Validation activity — e.g., "Present a pricing page and ask for a letter of intent or pilot commitment from 2 companies"]

Signal that confirms the idea: [What you need to see to move forward with confidence]
Signal that kills the idea: [What you need to see to stop and pivot]

How to Use

  1. Describe your background with specificity. "I worked in healthcare" produces generic ideas. "I spent 8 years managing clinical trial operations at a CRO, specifically focused on site activation and regulatory submission timelines" produces ideas that are actionable and defensible.
  2. The "observed problems" field is often the most valuable. Include problems you have personally experienced, seen clients struggle with, or heard repeatedly in your professional network. These are the signals that a market exists.
  3. Be honest about your constraints. A solo founder with no venture funding and six months of runway needs a different idea than a team of three with €500K. The prompt will calibrate accordingly.
  4. Take the defensibility ratings seriously. A High severity problem with Low defensibility is a dangerous place to build — you may validate a real market only to watch a larger competitor copy your approach with more resources.
  5. Run the validation roadmap before writing a line of code or signing a lease. The fastest path to knowing whether an idea is real is talking to ten potential customers with a specific hypothesis in hand.

Example Input

## Information about me

- My professional background: 9 years in commercial real estate — 4 as a broker, 5 as Head of Acquisitions at a mid-sized property investment firm in Germany. Left 6 months ago to explore building something.
- Skills: Deep industry relationships, financial modeling, investment analysis, deal structuring. Technically non-technical (no coding) but comfortable with no-code tools and APIs.
- Industries I understand from the inside: Commercial real estate investment and brokerage in the DACH region, particularly office and logistics assets
- Preferred business model: Would prefer SaaS or a productized service — want something scalable. Open to starting with services and transitioning to software.
- Resources: 12 months of personal runway, no co-founder yet, no institutional funding. Willing to raise a small pre-seed if the idea warrants it.
- Constraints: Must target B2B. Prefer Germany/DACH market initially. Need to reach first revenue within 6 months.
- Observed problems: Investment analysts at property firms spend 60–70% of their time manually collecting and structuring data for investment memos — from PDFs, broker emails, and market reports — before any actual analysis begins. The data gathering is the bottleneck, not the thinking.

Tips

  • The "observed problems" field is worth ten minutes of dedicated thinking before you run the prompt. Sit quietly and write down every operational frustration you have seen in your industry that still has no good solution. That list is a business idea backlog. The prompt will help you evaluate which ones are worth pursuing.
  • Rate the AI fit honestly. The most common failure mode in AI business ideation is overestimating how much AI adds. If the product would be nearly as useful without the AI component, you're building software — which is fine, but the AI is not the differentiator you think it is.
  • Defensibility is the criterion most people underweigh at the idea stage. A good problem with a clear AI solution is not a business if any well-funded team could replicate it in three months. Your defensibility comes from data, distribution, relationships, or switching costs — identify which one your idea relies on.
  • The validation roadmap is more valuable than the idea list. Anyone can generate ideas. The discipline is in testing them cheaply and quickly before committing resources. The "wizard of oz" approach — manually delivering the AI's output before building automation — is almost always the fastest way to validate.
  • Talk to customers before you talk to investors. The strongest position when fundraising is ten customer conversations that confirm the problem and two letters of intent from companies willing to pilot. Ideas without that evidence are worth very little, regardless of how well you can pitch them.

We use cookies

We use cookies to ensure you get the best experience on our website. For more information on how we use cookies, please see our cookie policy.

By clicking "Accept", you agree to our use of cookies.
Learn more.