startup idea

**Startup validation follows a problem-first approach** Startup validation follows a problem-first approach: identify a real problem, validate with potential users, then build a minimal viable product (MVP) to test the solution. AI enables new product categories but product-market fit remains the fundamental requirement for success. Problem identification: start with a pain point, not a technology; "what problem can AI solve?" not "what can I build with AI?" Customer discovery interviews validate that the problem exists and matters. User validation: talk to potential customers before building; understand their current solutions, willingness to pay, and urgency of the problem. Beware of false positives from polite feedback. MVP principles: build the smallest thing that tests your core hypothesis; for AI products, this might be a Wizard-of-Oz prototype (human-powered initially) or a limited-scope model. Iterate quickly: launch early, gather feedback, and refine. AI-specific considerations: data availability (do you have or can you get training data?), technical feasibility (can AI actually solve this?), and differentiation (what's your moat?). Common mistakes: building before validating, falling in love with technology over problem, and underestimating go-to-market. Problem-market fit precedes product-market fit.