Find Ai Startup Ideas

Workflow

1. Inventory the founder's edge: prior roles, rare intersections, frontier tools, communities, and non-obvious access.
2. List painful workflows they have seen directly. Prefer repetitive clerical, outsourced, consultant-heavy, QA, search, data, billing, logistics, or coordination work that AI can now automate or compress.
3. Run the "if not us, then who?" test. Favor ideas where the founder has a credible right to win because they understand the workflow better or can access it faster than outsiders.
4. If the founder lacks a strong internal edge, get them outside the house: shadow operators, use family or friend access, do ride-alongs, join practitioner communities, or take short-term operational work to learn the workflow from the inside.
5. Revisit niches that used to look too small. With AI replacing labor, narrow workflows can now support meaningful businesses.
6. Filter each candidate idea by pain intensity, access to users, why-now from current AI capabilities, and speed of validation.
7. Ignore weak external signals. Do not let TechCrunch competitors, investor skepticism, or doomscrolling outweigh direct evidence from users and working prototypes.
8. Return a ranked opportunity map and the next fieldwork steps needed to validate the top one or two ideas.

Output

• A concise idea memo or pivot recommendation with sections for Founder edge, Target user and painful workflow, Why now with AI, Why this team could win, Fastest path to validate, and Reasons to reject or de-prioritize.

Quality Bar

• Prefer operational guidance over abstract advice.
• Keep the response grounded in the source thesis.
• Ask follow-up questions if key inputs are missing.
• Do not repeat transcript wording unless a short quote is necessary.
• Prefer ideas tied to real access over random category brainstorms.
• Distinguish direct evidence from speculation.
• If the founder lacks access, return a field-research plan before suggesting a final idea.
• Do not recommend deceptive or non-consensual research tactics.

References

• Read references/source-context.md for thesis, triggers, and non-goals.
• Read references/transcript.md for source evidence and timestamps.
• Read references/patterns.md for concrete anti-patterns, fixes, and implementation details.