What is Real Problem AI?

Real Problem AI is a curated directory of 100 real problems worth building an AI startup around. Each problem is scored honestly on severity, AI feasibility, market signal, and competition gap. We're for builders looking for what to build, not for ethics debates about AI.

What is the difference between Real Problem AI and AI ethics or AI risks indexes?

Real Problem AI catalogues problems AI can solve — productive, commercial opportunities for builders. AI ethics indexes catalogue problems with AI itself (bias, environmental cost, copyright, consciousness). Different intent, different audience: founders and indie hackers vs researchers and policy folks.

What should I build with AI in 2026?

Look for problems with high severity (it costs people hours, sleep, or revenue every week), high AI feasibility (LLMs, vision or voice are great at this today), strong market signal (Reddit threads + willingness to pay), and a clear competition gap (incumbents miss the same obvious thing). Real Problem AI scores 100 such problems for you on those four axes.

Where do Real Problem AI's startup ideas come from?

From three sources, scanned each cycle: 1) Reddit threads where people describe a real friction in their own words, 2) podcasts and talks of founders young builders trust (Nikhil Kamath, Aman Gupta, Ankur Warikoo, Raj Shamani, Andrej Karpathy, Greg Isenberg), and 3) a continuous 1,000+ source scan covering forums, app store reviews, regulator filings, and field interviews.

Is Real Problem AI free?

Yes. Browsing every problem, opening every score breakdown, raising your hand on a co-founder match, and the founder vault are all free. There is no paywall.

How is each problem scored?

Four axes, each 1-10, weighted: Problem Severity (30%), AI Feasibility (25%), Market Signal (25%), Competition Gap (20%). Weighted average is the Opportunity Score on each card. Only problems clearing 7.0+ make the live list.

Why are library docs still written for humans when the first reader is now an agent?

Cursor / Claude Code / Devin are now the primary reader of any new SDK; doc sites front-load tutorials, marketing, navigation chrome and screenshots that waste agent context window and degrade tool-call accuracy.

Category: Others · Trend: LLM · Opportunity score: 8.3 / 10

What is the “Why are library docs still written for humans when the first reader is now an agent?” problem in 2026?

Who has this problem?

AI-native developers using agentic coding (Cursor, Claude Code, Devin, Copilot) on libraries and APIs that haven't shipped agent-grade docs.

Evidence this problem is real

“Why are people still telling me what to do? I don't want to do anything. What is the thing I should copy-paste to my agent?”

Sourced from Andrej Karpathy, Sequoia Ascent 2026 summary (link)

Existing players in this space

Mintlify / ReadMe / GitBook — Beautiful human docs; AI-ready is a chatbot bolt-on, not structured agent contract.
Context7 — Aggregates llms.txt at consumer side; doesn't fix source library docs.
llms.txt standard — Convention with no opinionated tooling, validation, or coverage SLA.
Anthropic docs MCP / OpenAI cookbook — First-party only.

What existing players are missing

Stripe Atlas for agent-grade docs: ingest a repo + OpenAPI spec, output a versioned, tested, agent-shaped doc bundle the maintainer can host. Per-library SaaS to maintainers ($99-499/mo).

How Real Problem AI scores this opportunity

Aggregate score: 8.3 / 10. Four-axis rubric:

Problem severity: 8 / 10
AI feasibility today: 9 / 10
Market signal: 8 / 10
Competition gap: 8 / 10

How to build a solution: stack hints

OpenAPI + JSDoc / TSDoc parser
Tested example generation
llms.txt + MCP bundle builder
Per-version diff + regression checks