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 do Cursor and Claude Code bill me $1,500 a month without telling me which feature burned it?

Solo and small-team devs blow $500-$2000/mo on AI coding subscriptions with zero per-feature, per-PR or per-repo attribution.

Category: Others · Trend: Agent · Opportunity score: 8.5 / 10

What is the “Why do Cursor and Claude Code bill me $1,500 a month without telling me which feature burned it?” problem in 2026?

Solo and small-team devs blow $500-$2000/mo on AI coding subscriptions with zero per-feature, per-PR or per-repo attribution.

Who has this problem?

Indie hackers, contractors, and 2-5 person engineering teams paying for Cursor / Claude Code / Cody / Copilot.

Evidence this problem is real

“I just got a $1,800 Cursor invoice and I have absolutely no idea which 3 days of which 2 features ate the budget.”

Sourced from Hacker News + Cursor Discord recurring threads, 2026 Q1-Q2

Existing players in this space

Cursor billing dashboard — Aggregate counters; no per-feature breakdown.
Helicone / Langfuse — For API-direct usage; do not see Cursor / IDE-mediated calls.

What existing players are missing

An IDE side-loaded observer that maps every token spend to feature, file, branch and PR. Sells by saving 30%+ of the bill in week 1.

How Real Problem AI scores this opportunity

Aggregate score: 8.5 / 10. Four-axis rubric:

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

How to build a solution: stack hints

VS Code / Cursor extension
Local SQLite event store
Stripe webhook for invoice reconciliation
Slack/email weekly digest